{"schema_version":"onlylabs.public_analysis_evidence.v1","title":"AI21 Labs analysis evidence pack","description":"Public onlylabs evidence pack for cited agent analysis: captured pages, ranked public signals, and stored web-search provenance used by the background analysis workflow.","url":"https://onlylabs.fyi/labs/ai21","json_url":"https://onlylabs.fyi/analysis/ai21/evidence.json","generated_at":"2026-06-11T15:10:41.764Z","org":{"slug":"ai21","name":"AI21 Labs","category":"neolab","category_label":"Neolab","dossier_url":"https://onlylabs.fyi/labs/ai21"},"analysis":null,"workflow":{"version":"onlylabs-deepagents-analysis-v3","provider":null,"model":null,"agent":null,"public_pack_mode":"local-pages-and-events","live_web_fetches":false,"note":"Public evidence exports do not trigger live Exa calls; stored Exa provenance is included when analysis metadata contains it."},"stats":{"pages":28,"events":58,"web":0,"evidence":86,"signal_desks":{"hiring":0,"forks":6,"releases":31,"talking":0,"repos":21},"data_radar_lanes":null,"data_radar_matches":null,"stored_analysis_evidence":null,"stored_analysis_web":null,"stored_analysis_signal_desks":null,"stored_analysis_data_radar_lanes":null,"stored_analysis_data_radar_matches":null},"stored_web_provenance":null,"evidence":[{"ref":"P1","kind":"page","title":"AI21Labs/sense-bert repository metadata","date":"2026-06-11T04:20:02.323328+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/sense-bert","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/sense-bert\n\nDescription: This is the code for loading the SenseBERT model, described in our paper from ACL 2020.\n\nLanguage: Python\n\nLicense: Apache-2.0\n\nStars: 48\n\nForks: 11\n\nOpen issues: 5\n\nCreated: 2020-07-06T07:54:34Z\n\nPushed: 2023-03-24T23:47:37Z\n\nDefault branch: master\n\nFork: no\n\nArchived: yes\n\nREADME:\n# SenseBERT: Driving Some Sense into BERT\n\nThis is the code for loading the SenseBERT model, described in [our paper](https://www.aclweb.org/anthology/2020.acl-main.423.pdf) from ACL 2020.\n\n## Available models\n\nWe made two SenseBERT models public:\n\n* sensebert-base-uncased\n* sensebert-large-uncased\n\nThese models have the same number of parameters as Google's BERT models, except for the following (both changes are described in our paper thoroughly):\n1. We use a larger vocabulary.\n2. We add a supersense prediction head. The sense embeddings are also used as inputs to the model. \n\n## Requirements\n\n* Python 3.7 or higher\n* TensorFlow 1.15\n* NLTK\n\nYou can install these using:\n```bash\npip install -r requirements.txt\n```\n\n## Usage\n\n### Supersense and MLM predictions\n\nThis is an example for making Masked Language Modeling (MLM) and supersense predictions based on SenseBERT:\n```python\nimport tensorflow as tf\nfrom sensebert import SenseBert\n\nwith tf.Session() as session:\nsensebert_model = SenseBert(\"sensebert-base-uncased\", session=session) # or sensebert-large-uncased\ninput_ids, input_mask = sensebert_model.tokenize([\"I went to the store to buy some groceries.\", \"The store was closed.\"])\nmodel_outputs = sensebert_model.run(input_ids, input_mask)\n\ncontextualized_embeddings, mlm_logits, supersense_logits = model_outputs # these are NumPy arrays\n```\nNote that both vocabularies (tokens and supersenses) are available for you via ```sensebert_model.tokenizer```. For example, in order to predict the supersense of the word 'groceries' in the above example, you may run\n```python\nimport numpy as np\n\nprint(sensebert_model.tokenizer.convert_ids_to_senses([np.argmax(supersense_logits[0][9])]))\n```\nThis will output:\n```\n['noun.artifact']\n```\n### Fine-tuning\n\nIf you want to fine-tune SenseBERT, run\n```python\nsensebert_model = SenseBert(\"sensebert-base-uncased\", session=se"},{"ref":"P2","kind":"page","title":"AI21Labs/pmi-masking repository metadata","date":"2026-06-11T04:20:02.195235+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/pmi-masking","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/pmi-masking\n\nDescription: This repository includes the masking vocabulary used in the ICLR 2021 spotlight PMI-Masking paper\n\nStars: 14\n\nForks: 3\n\nOpen issues: 1\n\nCreated: 2021-07-07T13:35:35Z\n\nPushed: 2021-08-09T14:03:05Z\n\nDefault branch: main\n\nFork: no\n\nArchived: yes\n\nREADME:\n# PMI-Masking\nThis repository includes the list of masked spans (the masking vocabulary) that is used in the [ICLR 2021 spotlight](https://iclr.cc/virtual/2021/spotlight/3496) PMI-Masking [paper](https://openreview.net/forum?id=3Aoft6NWFej), overviewed in this [blogpost](https://www.ai21.com/blog/pmi-masking). Below,\n\n* Section 1 provides the list construction details. \n* Section 2 details the PMI-Masking method for bidirectional masked language models, which essentially treats all of the spans in the provided lists as units for masking. \n* Section 3 provides some scores of recently trained models on the [SQuAD2.0](https://rajpurkar.github.io/SQuAD-explorer/) and [RACE](https://www.cs.cmu.edu/~glai1/data/race/) benchmarks.\n\n## Section 1: How we constructed the masking vocabulary\n\nGiven a pretraining corpus, we use its n-gram co-occurrence statistics to compile our masking vocabulary. We consider word n-grams of lengths 2–5 having over 10 occurrences in the corpus, and include the highest ranking collocations in the corpus, as measured via our proposed PMI_n measure (see equation below). Noticing that the PMI_n measure is sensitive to the length of the n-gram, we assemble per-length rankings for each n ∈ {2, 3, 4, 5}, and integrate these rankings to compose the masking vocabulary according to the following policy: 50% of the assembled list are the top ranking bigrams, 25% are the top ranking trigrams, and 12.5% are the top ranking 4-grams and 5-grams. After conducting a preliminary evaluation of how an n-gram’s quality as a collocation degrades with its PMI_n rank (detailed in the appendix of the paper), we chose the masking vocabulary size to be 800K, for which approximately half of pretraining corpus tokens were identified as part of some correlated n-gram\n\nThe PMI-Masking lists included in this repo are attained by using the following ranking score:\n\n<img src=\"https://render.g"},{"ref":"P3","kind":"page","title":"AI21Labs/jurades repository metadata","date":"2026-06-11T04:20:02.037705+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/jurades","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/jurades\n\nDescription: AI Charades game\n\nLanguage: JavaScript\n\nLicense: Apache-2.0\n\nStars: 3\n\nForks: 2\n\nOpen issues: 0\n\nCreated: 2021-08-04T18:16:41Z\n\nPushed: 2021-08-11T13:12:24Z\n\nDefault branch: master\n\nFork: no\n\nArchived: yes\n\nREADME:\n<img width=\"350\" align=\"right\" src=\"https://user-images.githubusercontent.com/7150767/128235508-b8ec81f5-874d-4496-b6d0-e7bdbf3f7017.png\">\n<img width=\"300\" alt=\"Jurades\" src=\"https://user-images.githubusercontent.com/7150767/128235107-903820d0-dbf4-4d73-98d8-64542298b02f.png\">\n\n## An AI Charades game, _powered by Jurassic-1 language model by AI21Labs Studio_\n\n- [Running the game](#running-the-game)\n- [Server](#server)\n- [Client](#client)\n- [How to play](#how-to-play)\n- [Configuration](#configuration)\n\n# Running the game\n\nThe game consist of:\n- A server app, which produces the guesses by sending requests to AI21 Studio. This is done from the backend in order to keep the AI21 Studio API key secured.\n- A client web app which is the actual game, sending request to the server mentioned above.\n\n## Server\n\nYou need to have `python > 3.5` and `pip` to install the dependencies and run the server. \n\nIn order to run the server locally:\n\n1. Enter the server folder and install the python dependencies using pip\n```shell\ncd server\npip install -r requirements.txt\n```\n2. Set your AI21 Studio API key as an environment variable.\n_(Replace <YOUR-API-KEY> with the key you generated in AI21 Studio)_\n```shell\nexport AI21_API_KEY=<YOUR-API-KEY>\n```\n3. Start Flask's development server\n```shell\npython main.py\n```\n***🎉 The server should be up and running locally on port 5000.***\n\n## Client\n\nYou need to have `node` and `npm` installed for running and building the client app.\n\nFirst of all, make sure to enter the client app's folder:\n```shell\n# Return to the repo's parent folder if you were inside the server folder\ncd ..\n# Enter the client app folder\ncd client\n```\nInstall the relevant packages\n```shell\nnpm i\n```\n\n### Running the local dev version\nSimply run the following inside the client folder\n```shell\nnpm start\n```\nAnd the app will be accessible at `http://localhost:3000`\n\n### Building for production\nRun\n```shell\nnpm run build\n```\nA production"},{"ref":"P4","kind":"page","title":"AI21Labs/jurassic-chess repository metadata","date":"2026-06-11T04:20:01.99702+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/jurassic-chess","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/jurassic-chess\n\nDescription: A talking chess AI\n\nLanguage: Python\n\nLicense: Apache-2.0\n\nStars: 5\n\nForks: 5\n\nOpen issues: 0\n\nCreated: 2021-08-10T06:51:54Z\n\nPushed: 2021-08-11T13:15:17Z\n\nDefault branch: master\n\nFork: no\n\nArchived: yes\n\nREADME:\n# A talking chess AI\n\n## _powered by Jurassic-1 language model by AI21Labs Studio_\n\n- [Running the game](#running-the-game)\n- [How to play](#how-to-play)\n- [Configuration](#configuration)\n\n# Running the game\n\nThe game runs as a local web server. To play you need to run the server and browse to http://127.0.0.1:5000/.\nThe server will automatically move the pieces for the AI and give vocal feedback to the user.\n\nYou need to have `python > 3.7` and `pip` to install the dependencies and run the server. \n\nIn order to run the server:\n\n1. Install the python dependencies using pip:\n```shell \npip install -r requirements.txt\n```\nIf you are running on Windows and the speech doesn't work, try to edit requirement.txt and change the version of pyttsx3 to 2.71.\n\n2. Give running permission to the chess engine:\n```shell\nchmod +x engines/stockfish\n```\n3. Edit config.ini and put your AI21 Studio API key in the reserved place.\n4. Start Flask's development server:\n```shell\npython play.py\n```\n***🎉 The server should be up and running locally on port 5000.***\n\n# How to play?\n\n- A new game starts immediately when the server runs. Human plays white. \n- Browse to http://127.0.0.1:5000/\n- Click on a piece to move, then click on a target square. The AI makes its response automatically.\n- From time to time the AI will give vocal feedback to the human moves. At the top of the board you can select the tone of the feedback (tutor or taunt)\n\n# Configuration\n\nYou can change a few of the game's configuration parameters by editing config.ini:\n\n1. AI21 studio API parameters.\n```ini\napi_key = <<your api key>>\nurl = https://api.ai21.com/studio/v1/j1-jumbo/complete \n#url = https://api.ai21.com/studio/v1/j1-large/complete\n```\n- `api_key` You AI21 studio API key.\n- `url` The API url address. By default we use the url to J1-Jumbo model. Replace the mask to use J1-Large instead.\n\n2. Speech parameters.\n```ini\nbad_move_threshold = 50\ngood_move_threshold = 10\n"},{"ref":"P5","kind":"page","title":"AI21Labs/lm-evaluation repository metadata","date":"2026-06-11T04:20:01.984942+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/lm-evaluation","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/lm-evaluation\n\nDescription: Evaluation suite for large-scale language models.\n\nLanguage: Python\n\nLicense: Apache-2.0\n\nStars: 130\n\nForks: 15\n\nOpen issues: 2\n\nCreated: 2021-08-05T13:22:20Z\n\nPushed: 2021-08-15T13:49:52Z\n\nDefault branch: main\n\nFork: no\n\nArchived: yes\n\nREADME:\n# LM Evaluation Test Suite\nThis repo contains code for running the evaluations and reproducing the results from the [Jurassic-1 Technical Paper](https://uploads-ssl.webflow.com/60fd4503684b466578c0d307/61138924626a6981ee09caf6_jurassic_tech_paper.pdf) (see [blog post](https://www.ai21.com/blog/announcing-ai21-studio-and-jurassic-1)), with current support for running the tasks through both the [AI21 Studio API](https://studio.ai21.com/) and [OpenAI's GPT3 API](https://beta.openai.com/).\n\n## Citation\nPlease use the following bibtex entry:\n```\n@techreport{J1WhitePaper,\nauthor = {Lieber, Opher and Sharir, Or and Lenz, Barak and Shoham, Yoav},\ntitle = {Jurassic-1: Technical Details And Evaluation},\ninstitution = {AI21 Labs},\nyear = 2021,\nmonth = aug,\n}\n```\n\n## Installation\n```\ngit clone https://github.com/AI21Labs/lm-evaluation.git\ncd lm-evaluation\npip install -e .\n```\n\n## Usage\nThe entry point for running the evaluations is lm_evaluation/run_eval.py, which receives a list of tasks and models to run. \n\nThe models argument should be in the form \"provider/model_name\" where provider can be \"ai21\" or \"openai\" and the model name is one of the providers supported models.\n\nWhen running through one of the API models, set the your API key(s) using the environment variables AI21_STUDIO_API_KEY and OPENAI_API_KEY. Make sure to consider the costs and quota limits of the models you are running beforehand.\n\nExamples:\n```console\n# Evaluate hellaswag and winogrande on j1-large\npython -m lm_evaluation.run_eval --tasks hellaswag winogrande --models ai21/j1-large\n\n# Evaluate all multiple-choice tasks on j1-jumbo\npython -m lm_evaluation.run_eval --tasks all_mc --models ai21/j1-jumbo\n\n# Evaluate all docprob tasks on curie and j1-large\npython -m lm_evaluation.run_eval --tasks all_docprobs --models ai21/j1-large openai/curie\n\n```\n\n## Datasets\nThe repo currently support the zero-shot multiple-choice and document"},{"ref":"P6","kind":"page","title":"AI21Labs/MRKL_synthetic_data repository metadata","date":"2026-06-11T04:20:01.141405+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/MRKL_synthetic_data","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/MRKL_synthetic_data\n\nDescription: All synthetic data used for the experiments described in MRKL white paper https://www.ai21.com/blog/jurassic-x-crossing-the-neuro-symbolic-chasm-with-the-mrkl-system.\n\nLicense: Apache-2.0\n\nStars: 5\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2022-04-27T06:51:01Z\n\nPushed: 2022-05-01T06:51:08Z\n\nDefault branch: main\n\nFork: no\n\nArchived: yes\n\nREADME:\n# MRKL_synthetic_data\nThis repository contains all of the artificial data used for the experiments described in MKRL white paper:\nhttps://www.ai21.com/blog/jurassic-x-crossing-the-neuro-symbolic-chasm-with-the-mrkl-system\n\nThe repository is divided into folders, containing the neccassiry data to replicate the experiments described in the white paper, and are numbered accordingly.\n\nEach data file is in jsonlines format, containing the columns: \n- **prompt**: the text to be evaluated by the model.\n- **completion**: the correct text prediction for the model.\n- **execution**: the correct evaluation of the completion formula done by a calculator.\n\nExample:\n{\"prompt\":\"How much is 27395 plus 87971 plus 25658?\",\"execution\":141024.0,\"completion\":\"X=(27395+87971+25658)\"}\n\nSome of the files contain extra columns which may be used to further stratify the data. Those columns are not required to replicate any of the experiments desribed in the white paper, but enable a more elaborate analysis."},{"ref":"P7","kind":"page","title":"AI21Labs/dev-envs repository metadata","date":"2026-06-11T04:20:00.946084+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/dev-envs","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/dev-envs\n\nLanguage: Shell\n\nStars: 1\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2022-12-20T11:48:51Z\n\nPushed: 2022-12-20T11:49:15Z\n\nDefault branch: master\n\nFork: no\n\nArchived: yes\n\nREADME:\n# Development Environment and Experience\n\nA collection of [helper scripts](#helpers) to bootstrap your Mac box for development.\nLet's start...\n\n---\n\n**If you find any issue or you feel something is missing, we encourage you to contribute to the project.** For more information, see [contributing](#contributing).\n\n---\n\n## Disclaimer\n\n1. **We do not modify user settings automatically** — to complete the setup, please complete the installation as noted at the first line of each language below\n2. **Always start by bootstrapping your machine.** For more information, see [Getting Started](#getting-started)\n\n## Getting Started\n\n1. Clone the repository, **or**\n\n```shell\ngit clone git@bitbucket.org:ai21labs/dev-envs.git\n```\n\n2. [Download it as a zip ball](https://bitbucket.org/ai21labs/dev-envs/src/master/), unzip to a directory and change into it (if you are a new on the crew, you probably want this)\n\n```shell\nunzip -q -o ai21-dev-envs-*.zip\ncd ai21-dev-envs-*/\n```\n\n## Bootstrap Your Mac\n\n1. [Kandji](https://www.kandji.io/mdm/start/product/) should have completed bootstrapping your Mac installation with [brew](https://brew.sh/) as described [here](https://ai21labs.atlassian.net/wiki/spaces/engineering/pages/155812007/How+to+Bootstrap+your+Mac+with+Kandji)\n- Verify by running `brew doctor`\n- If you find some packages are missing, feel free to contribute to [infra-mac-bootstrap](https://bitbucket.org/ai21labs/infra-mac-bootstrap/src/master)\n2. Our [helper scripts](#helpers) require Bash version 5.1.8 (or higher). By default MacOS comes with Bash version 3.2.x. Refer to [this section](https://itnext.io/upgrading-bash-on-macos-7138bd1066ba) to set Bash version 5.x.x as default\n\n## Generate SSH keys\n\n1. Execute `./helpers/ssh-key-gen.sh`\n2. Store the generated SSH keys in your password manager\n\n## GitHub\n\n1. Execute `./helpers/config-github.sh`\n2. Authorize [an SSH key for use with SAML single sign-on](https://docs.github.com/en/github/authenticating-to-github/authenticating-with-saml-si"},{"ref":"P8","kind":"page","title":"AI21Labs/SageMaker repository metadata","date":"2026-06-11T04:20:00.892991+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/SageMaker","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/SageMaker\n\nDescription: Examples for using AI21's models through Amazon SageMaker\n\nLanguage: Jupyter Notebook\n\nStars: 32\n\nForks: 16\n\nOpen issues: 1\n\nCreated: 2022-11-22T14:42:37Z\n\nPushed: 2024-12-01T12:15:11Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n# Deploy AI21 Labs Language Models on AWS SageMaker\n\nExamples for using Jurassic-2 models through Amazon SageMaker.\n\n## Installation\n\nYou can install the package using pip:\n\n```\npip install -U \"ai21[AWS]\"\n```\n\nAnd then import the package:\n\n```\nimport ai21\n```\n\n## Quickstart\n\nIn order to use this library, you must configure your AWS credentials.\n\nAll set up? Make your first interaction with Jurassic-2:\n\n```python\nresponse = ai21.Completion.execute(destination=ai21.SageMakerDestination(\"j2-ultra\"),\nprompt=\"To be, or\",\nmaxTokens=4,\ntemperature=0)\n\nprint(response['completions'][0]['data']['text'])\n# not to be: that is the question\n```\n\nFor more comprehensive quickstart guides, see:\n\n#### Foundation models\n\n[Jurassic-2 Ultra](J2_Ultra_example_model_use.ipynb)\n\n[Jurassic-2 Mid Instruct](J2_Mid_example_model_use.ipynb)\n\n[Jurassic-2 Light](J2_Light_example_model_use.ipynb)\n\n### Task-specific models\n\n[AI21 Summarize](AI21_Summarize_example_model_use.ipynb)\n\n[AI21 Contextual Answers](AI21_ContextualAnswers_example_model_use.ipynb)\n\n[AI21 Paraphrase](AI21_Paraphrase_example_model_use.ipynb)\n\n[AI21 Grammatical Error Correction (GEC)](AI21_GEC_example_model_use.ipynb)\n\n## Learn more\n\n- [AI21 Studio website](http://www.ai21.com/studio)\n\n- [Our technical docs](http://docs.ai21.com)\n\n- [AI21 Studio playground](https://studio.ai21.com/playground) - no code environment to test and perfect your prompts"},{"ref":"P9","kind":"page","title":"AI21Labs/github-migration repository metadata","date":"2026-06-11T04:20:00.866003+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/github-migration","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/github-migration\n\nLanguage: JavaScript\n\nStars: 1\n\nForks: 1\n\nOpen issues: 2\n\nCreated: 2022-12-20T10:55:34Z\n\nPushed: 2025-02-15T00:08:31Z\n\nDefault branch: master\n\nFork: no\n\nArchived: no\n\nREADME:\n# Github Migration\n\nMigrate Bitbucket repositories to Github\n\n## Prerequisites\n\n- Install nvm\n\n```shell\nbrew install nvm\n```\n\n- Configure Node Version Manager (nvm) as described [here](https://github.com/AI21Labs/dev-envs/blob/master/docs/nodejs-dev.md#configure-version-manager)\n\n- Configure SSH key in GitHub by clonning [dev-envs](https://github.com/AI21Labs/dev-envs) and running\n[config-github.sh script](https://github.com/AI21Labs/dev-envs#github)\n\n## Getting Started\n\n### Install deps\n\n```bash\nsource ./init.sh\n```\n\n### Github Mobile\n\n- Download [GitHub app for mobile](https://github.com/mobile)\n\n### Create Access Tokens\n\n- [Create Github Personal Access Token (PAT) with the scopes specified](https://docs.github.com/en/early-access/enterprise-importer/preparing-to-migrate-with-github-enterprise-importer/managing-access-for-github-enterprise-importer#creating-a-personal-access-token-for-github-enterprise-importer)\n- [Create Bitbucket Access Token](https://confluence.atlassian.com/bitbucketserver072/personal-access-tokens-1005335924.html#:~:text=To%20generate%20a%20personal%20access,the%20users%20personal%20tokens%20page.&text=Use%20permissions%20to%20get%20the%20correct%20access%20for%20different%20users.)\n- Update the following in [.env-template](./.env-template) file:\n- GitHub username\n- GitHub and BitBucket tokens\n- Repository to migrate\n\n## Migrate a Bitbucket Repository\n\n- Run migration script\n\n```bash\n$ node migration.js --help\nUsage: migration.js --ght [string] --bbt [string] --repo [string] --owner [string]\n\nOptions:\n--help Show help [boolean]\n--version Show version number [boolean]\n--ght Github Personal Access Token (PAT) [require]\n--bbt BitBucket Personal Access Token (PAT) [required]\n--repo Repository to migrate [required]\n--owner Github username [required]\n```\n- For example:\n\n```bash\nnode migration.js --bbt BB_TOKEN --ght GH_TOKEN --owner USER --repo REPO_NAME\n```\n\n- Update Cloudbuild with the new github repository\n- [connect repository](https://con"},{"ref":"P10","kind":"page","title":"AI21Labs/Rick-and-Morty-episode-generator repository metadata","date":"2026-06-11T04:20:00.788891+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/Rick-and-Morty-episode-generator","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/Rick-and-Morty-episode-generator\n\nDescription: An interactive one-pager browser game, primarily built with React.js. It makes use of the state-of-the-art Jurassic-2 language models to generate made-up episodes of the popular animated science fiction TV show Rick and Morty.\n\nLanguage: JavaScript\n\nLicense: MIT\n\nStars: 16\n\nForks: 0\n\nOpen issues: 0\n\nCreated: 2023-01-12T10:32:39Z\n\nPushed: 2023-09-13T07:42:41Z\n\nDefault branch: main\n\nFork: no\n\nArchived: yes\n\nREADME:\n<p align=\"center\">\n<img src=\"https://user-images.githubusercontent.com/118735447/212035080-e4f736c7-94a7-4381-a3c6-ce18cdb7e761.png\" width=\"400\">\n</p>\n\n# Rick and Morty Episode Generator\n\nAn interactive one-pager browser game, primarily built with React.js. It makes use of the state-of-the-art Jurassic-2 language models to generate made-up episodes of the popular animated science fiction TV show Rick and Morty.\n\n## Getting Started\n1. Clone the repository \n\n`git clone https://github.com/AI21Labs/Rick-and-Morty-Episode-Generator.git`\n\n2. Navigate to cloned folder\n\n`cd Rick-and-Morty-Episode-Generator`\n\n3. Install dependencies\n\n`npm install`\n\n4. Start the game in your local environment\n\n`npm start`\n\n5. Open [http://localhost:3000](http://localhost:3000) to view it in the browser.\n\n## Play\n\n1. Get your API key from your [account](https://studio.ai21.com/account/account) page, and paste it in [apiKey.consts.js](src/api/constants/apiKey.consts.js) file. Now you are all set and ready to play!\n<br />\n\n2. Press the \"Generate Title\" button, which will invoke the generateTitle() function. This function sends an API request to the `/complete` endpoint of the Jurassic-2 Mid model. The main request includes a \"few shot\" `prompt` parameter with a value that prompts the model to generate a new made-up Rick and Morty episode title. Other parameters worth noticing are `maxTokens`, `temperature`, and `stopSequences`.\n<br />\n\n3. Once the model returns a response, it is rendered in the text field next to the \"Generate Title\" button.\n<br />\n\n4. The second button, \"Generate Plot\", becomes clickable, and when clicked, invokes the generatePlot() function. The function takes the generated title and uses it in its prompt v"},{"ref":"P11","kind":"page","title":"AI21Labs/ai21-tokenizer repository metadata","date":"2026-06-11T04:20:00.068417+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/ai21-tokenizer","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/ai21-tokenizer\n\nDescription: AI21's Jamba models tokenizers\n\nLanguage: Python\n\nLicense: Apache-2.0\n\nStars: 33\n\nForks: 4\n\nOpen issues: 1\n\nCreated: 2023-08-22T08:28:46Z\n\nPushed: 2025-10-27T23:44:08Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n<h1 align=\"center\">\n<a href=\"https://github.com/AI21Labs/ai21-tokenizer\">AI21 Labs Tokenizer</a>\n</h1>\n\n<p align=\"center\">\n<em>A SentencePiece based tokenizer for production uses with AI21's models</em>\n</p>\n\n<p align=\"center\">\n<a href=\"https://github.com/AI21Labs/ai21-tokenizer/actions?query=workflow%3ATest+event%3Apush+branch%3Amain\"><img src=\"https://github.com/AI21Labs/ai21-tokenizer/actions/workflows/test.yaml/badge.svg\" alt=\"Test\"></a>\n<a href=\"https://pypi.org/project/ai21-tokenizer\" target=\"_blank\"><img src=\"https://img.shields.io/pypi/v/ai21-tokenizer?color=%2334D058&label=pypi%20package\" alt=\"Package version\"></a>\n<a href=\"https://pypi.org/project/ai21-tokenizer\" target=\"_blank\"><img src=\"https://img.shields.io/pypi/pyversions/ai21-tokenizer?color=%2334D058\" alt=\"Supported Python versions\"></a>\n<a href=\"https://python-poetry.org/\" target=\"_blank\"><img src=\"https://img.shields.io/endpoint?url=https://python-poetry.org/badge/v0.json\" alt=\"Poetry\"></a>\n<a href=\"https://github.com/semantic-release/semantic-release\" target=\"_blank\"><img src=\"https://img.shields.io/badge/semantic--release-python-e10079?logo=semantic-release\" alt=\"Supported Python versions\"></a>\n<a href=\"https://opensource.org/licenses/Apache-2.0\" target=\"_blank\"><img src=\"https://img.shields.io/badge/License-Apache_2.0-blue.svg\" alt=\"License\"></a>\n</p>\n\n---\n\n## Prerequisites\n\n- If you wish to use the tokenizers for `Jamba Mini` or `Jamba Large`, you will need to request access to the relevant model's HuggingFace repo:\n- [Jamba Mini](https://huggingface.co/ai21labs/AI21-Jamba-Mini-1.6)\n- [Jamba Large](https://huggingface.co/ai21labs/AI21-Jamba-Large-1.6)\n\n## Installation\n\n### pip\n\n```bash\npip install ai21-tokenizer\n```\n\n### poetry\n\n```bash\npoetry add ai21-tokenizer\n```\n\n## Usage\n\n### Basic Usage\n\n```python\nfrom ai21_tokenizer import Tokenizer\n\n# Create tokenizer (defaults to Jamba Mini)\ntokenizer = Tokenizer.get_tokenizer()\n\n# Encode "},{"ref":"P12","kind":"page","title":"AI21Labs/in-context-ralm repository metadata","date":"2026-06-11T04:19:59.960918+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/in-context-ralm","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/in-context-ralm\n\nLanguage: Python\n\nLicense: Apache-2.0\n\nStars: 295\n\nForks: 28\n\nOpen issues: 4\n\nCreated: 2023-01-26T12:07:18Z\n\nPushed: 2023-12-20T07:55:54Z\n\nDefault branch: main\n\nFork: no\n\nArchived: yes\n\nREADME:\n# In-Context Retrieval-Augmented Language Models\n\nThis repo contains the code for reproducing the experiments on *WikiText-103* from [AI21 Labs](https://www.ai21.com/)' paper [In-Context Retrieval-Augmented Language Models](https://arxiv.org/abs/2302.00083) (In-Context RALM), to appear in the Transactions of the Association for Computational Linguistics (TACL).\n\nOur code is mainly based on the [Transformers](https://github.com/huggingface/transformers/) and [Pyserini](https://github.com/castorini/pyserini) libraries. \nWe test it on Python 3.8.\n\n## Table of Contents\n- [Setup](#setup)\n- [Retrieval](#retrieval)\n- [Evaluation](#evaluation)\n- [Language Models](#list-of-language-models)\n- [Evaluate models w/o retrieval](#evaluate-models-wo-retrieval)\n- [Evaluate models with retrieval](#evaluate-models-with-retrieval)\n- [Evaluate models with reranking](#reranking)\n- [Question Answering Experiments](#question-answering-experiments)\n- [Citation](#citation)\n\n## Setup\n\nTo install the required libraries in our repo, run:\n```bash\npip install -r requirements.txt\n```\nTo have a Pytorch version specific to your CUDA, [install](https://pytorch.org/) your version before running the above command.\n\n## Retrieval\n\n### BM25\n\nOur BM25 preparation script works with Pyserini, so Java 11 is required - see their [installation guide](https://github.com/castorini/pyserini/blob/master/docs/installation.md). \nIf you have Java 11 installed, make sure your `JAVA_HOME` environment variable is set to the correct path. \nOn a Linux system, the correct path might look something like `/usr/lib/jvm/java-11`. \nThen run:\n\n```bash\npython prepare_retrieval_data.py \\\n--retrieval_type sparse \\\n--tokenizer_name $MODEL_NAME \\\n--max_length 1024 \\\n--dataset_path wikitext \\\n--dataset_name wikitext-103-v1 \\\n--dataset_split [validation, test] \\\n--index_name wikipedia-dpr \\\n--forbidden_titles_path ralm/retrievers/wikitext103_forbidden_titles.txt \\\n--stride 4 \\\n--output_file $RETRIEVAL_FILE \\\n--num_"},{"ref":"P13","kind":"page","title":"AI21Labs/Parallel-Context-Windows repository metadata","date":"2026-06-11T04:19:59.959799+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/Parallel-Context-Windows","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/Parallel-Context-Windows\n\nLanguage: Python\n\nLicense: Apache-2.0\n\nStars: 107\n\nForks: 15\n\nOpen issues: 3\n\nCreated: 2023-02-22T13:14:24Z\n\nPushed: 2023-06-20T11:53:03Z\n\nDefault branch: main\n\nFork: no\n\nArchived: yes\n\nREADME:\n# Parallel Context Windows (PCW)\n\nThis repo contains the code for reproducing the classification experiments from [AI21 Labs](https://www.ai21.com/)' paper [Parallel Context Windows for Large Language Models\n](https://arxiv.org/abs/2212.10947). \nThe code was tested with python 3.10, for CPU, GPU and multiple GPU runs. Currently, the code supports using GPT2 and LLaMa model families.\n\n## Setup\n\nTo install the required libraries in our repo, run:\n```bash\npip install -r requirements.txt\n```\nTo have a Pytorch version specific to your CUDA, [install](https://pytorch.org/) your version before running the above command.\n\n## Evaluation\nDue to the fact that the paper's results were based on an earlier implementation of PCW and not [HuggingFace Transformers](https://huggingface.co/docs/transformers/index), the results produced using this code may differ slightly from those shown in the paper.\nTo reproduce similar results shown in the appendix for GPT2-XL for a specific dataset (for example SST2), simply run:\n```bash\npython run_evaluation.py \\\n--dataset sst2 \\\n--model gpt2-xl \\\n--n-windows 1 \\\n--n-windows 3 \\\n--subsample-test-set 250 \\\n--n-runs 30 \\\n--output-dir $OUTPUT_DIR\n```\nIn this run, PCW's performance is evaluated on a subsample (250 samples) of the full test set. \nThe experiment is repeated 30 times (with different random samples of training examples) for each number of windows (in this case - one and three). \nAs a default, the script uses as many examples per window as possible. \nNote that using a single window is equivalent to regular ICL settings. Thus, this run should give similar results to those shown in Table 5 for SST2 with GPT2-XL.\n\nThe evaluation output is a numpy file (shaped `[2,30]`) found in `$OUTPUT_DIR` with the mean accuracy for each repetition and number of windows.\nYou could read the file directly with np.load, or use utils.py function to load and plot the results.\nSee --help for further instructions.\n\n## PCW Usage exam"},{"ref":"P14","kind":"page","title":"AI21Labs/factor repository metadata","date":"2026-06-11T04:19:59.858372+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/factor","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/factor\n\nDescription: Code and data for the FACTOR paper\n\nLanguage: Python\n\nLicense: MIT\n\nStars: 53\n\nForks: 6\n\nOpen issues: 6\n\nCreated: 2023-07-13T12:15:34Z\n\nPushed: 2023-11-15T08:11:30Z\n\nDefault branch: main\n\nFork: no\n\nArchived: yes\n\nREADME:\n# FACTOR\n\nThis repo contains data from [AI21 Labs](https://www.ai21.com/)' paper [Generating Benchmarks for Factuality Evaluation of Language Models](https://arxiv.org/abs/2307.06908).\n\n## Data\n\nWe include the following FACTOR benchmarks for evaluating factuality of language models: \n* WIKI-FACTOR: Based on the Wikipedia section of [The Pile’s](https://github.com/EleutherAI/the-pile)) validation split. The dataset consists of 2994 examples.\n* NEWS-FACTOR: Based on Reuters articles extracted from [The RefinedWeb Dataset](https://arxiv.org/abs/2306.01116). The dataset consists of 1036 examples. \n* EXPERT-FACTOR: Based on the validation and test splits of [ExpertQA](https://arxiv.org/abs/2309.07852), a long-from question answering dataset. The benchmark consists of 236 examples.\n\n## Evaluation\n\n#### Setup\n\nTo install the required libraries in our repo, run:\n```bash\npip install -r requirements.txt\n```\nTo have a Pytorch version specific to your CUDA, [install](https://pytorch.org/) your version before running the above command.\n\n### List of Language Models\n\nIn the paper, we give the results for the following models (replace `$MODEL_NAME` with one of those). \n\n* GPT-2: `gpt2`, `gpt2-medium`, `gpt2-large`, `gpt2-xl`\n* GPT-Neo: `EleutherAI/gpt-neo-1.3B`, `EleutherAI/gpt-neo-2.7B`, `EleutherAI/gpt-j-6B`\n* OPT: `facebook/opt-125m`, `facebook/opt-350m`, `facebook/opt-1.3b`, `facebook/opt-2.7b`, `facebook/opt-6.7b`, `facebook/opt-13b`, `facebook/opt-30b`, `facebook/opt-66b`\n\n### Evaluation Script\n\nTo run evaluation on models over FACTOR datasets, please use the following command:\n```bash\npython python eval_factuality.py \\\n--data_file ./data/wiki_factor.csv \\\n--output_folder $OUTPUT_DIR \\\n--model_name $MODEL_NAME\n```\n\n## License\n* `wiki_factor`, `expert_factor` and code: Released under the MIT license.\n* `news_factor`: The [benchmark](data/news_factor.csv) is derived from [The RefinedWeb Dataset](https://arxiv.org/abs/2306.011"},{"ref":"P15","kind":"page","title":"AI21Labs/corefact repository metadata","date":"2026-06-11T04:19:59.777801+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/corefact","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/corefact\n\nStars: 0\n\nForks: 0\n\nOpen issues: 0\n\nCreated: 2023-07-10T12:35:36Z\n\nPushed: 2023-07-10T12:35:36Z\n\nDefault branch: main\n\nFork: no\n\nArchived: yes\n\nREADME: none published or not readable through the GitHub API."},{"ref":"P16","kind":"page","title":"AI21Labs/ai21-typescript repository metadata","date":"2026-06-11T04:19:59.17404+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/ai21-typescript","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/ai21-typescript\n\nDescription: AI21 Typescript SDK\n\nLanguage: TypeScript\n\nLicense: Apache-2.0\n\nStars: 13\n\nForks: 2\n\nOpen issues: 2\n\nCreated: 2024-11-05T09:44:25Z\n\nPushed: 2025-12-18T08:05:00Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n<h1 align=\"center\">\n<a href=\"https://github.com/AI21Labs/ai21-typescript\">AI21 Labs TypeScript SDK</a>\n</h1>\n\n<p align=\"center\">\n<a href=\"https://github.com/AI21Labs/ai21-typescript/actions/workflows/unittests.yml\"><img src=\"https://github.com/AI21Labs/ai21-typescript/actions/workflows/unittests.yml/badge.svg?branch=main\" alt=\"Test\"></a>\n<a href=\"https://github.com/AI21Labs/ai21-typescript/actions/workflows/integration-tests.yml\"><img src=\"https://github.com/AI21Labs/ai21-typescript/actions/workflows/integration-tests.yml/badge.svg?branch=main\" alt=\"Integration Tests\"></a>\n<a href=\"https://www.npmjs.com/package/ai21\" target=\"_blank\"><img src=\"https://img.shields.io/npm/v/ai21?color=%2334D058&label=npm%20package\" alt=\"Package version\"></a>\n<a href=\"https://nodejs.org/\" target=\"_blank\"><img src=\"https://img.shields.io/badge/node->=18.0.0-brightgreen\" alt=\"Supported Node.js versions\"></a>\n<a href=\"https://github.com/semantic-release/semantic-release\" target=\"_blank\"><img src=\"https://img.shields.io/badge/semantic--release-typescript-e10079?logo=semantic-release\" alt=\"Semantic Release Support\"></a>\n<a href=\"https://opensource.org/licenses/Apache-2.0\" target=\"_blank\"><img src=\"https://img.shields.io/badge/License-Apache_2.0-blue.svg\" alt=\"License\"></a>\n</p>\n\nThe AI21 API Client is a TypeScript library that provides a convenient interface for interacting with the AI21 API. It abstracts away the low-level details of making API requests and handling responses, allowing developers to focus on building their applications.\n\n- [Installation](#Installation) 💿\n- [Examples](#examples-tldr) 🗂️\n- [AI21 Official Documentation](#Documentation)\n- [Chat](#Chat)\n- [Maestro (Beta)](#Maestro)\n- [Conversational RAG (Beta)](#Conversational-RAG)\n- [Library](#Library)\n\n## Environment Support\n\nThis client supports both Node.js and browser environments:\n\n- **Node.js**: Works out of the box with Node.js >=18.0.0\n- **Browser**: Requires ex"},{"ref":"P17","kind":"page","title":"AI21Labs/ai21-python repository metadata","date":"2026-06-11T04:19:59.169383+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/ai21-python","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/ai21-python\n\nDescription: AI21 Python SDK\n\nLanguage: Python\n\nLicense: Apache-2.0\n\nStars: 70\n\nForks: 13\n\nOpen issues: 11\n\nCreated: 2023-12-10T09:20:15Z\n\nPushed: 2026-01-28T16:25:12Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n<h1 align=\"center\">\n<a href=\"https://github.com/AI21Labs/ai21-python\">AI21 Labs Python SDK</a>\n</h1>\n\n[//]: # \"Add when public\"\n[//]: # '<a href=\"https://github.com/AI21Labs/ai21/actions?query=workflow%3ATest+event%3Apush+branch%3Amain\"><img src=\"https://github.com/AI21Labs/ai21/actions/workflows/test.yaml/badge.svg\" alt=\"Test\"></a>'\n[//]: # '<a href=\"https://pypi.org/project/ai21\" target=\"_blank\"><img src=\"https://img.shields.io/pypi/pyversions/ai21?color=%2334D058\" alt=\"Supported Python versions\"></a>'\n\n<p align=\"center\">\n<a href=\"https://github.com/AI21Labs/ai21-python/actions/workflows/test.yaml\"><img src=\"https://github.com/AI21Labs/ai21-python/actions/workflows/test.yaml/badge.svg?branch=main\" alt=\"Test\"></a>\n<a href=\"https://github.com/AI21Labs/ai21-python/actions/workflows/integration-tests.yaml\"><img src=\"https://github.com/AI21Labs/ai21-python/actions/workflows/integration-tests.yaml/badge.svg?branch=main\" alt=\"Integration Tests\"></a>\n<a href=\"https://pypi.org/project/ai21\" target=\"_blank\"><img src=\"https://img.shields.io/pypi/v/ai21?color=%2334D058&label=pypi%20package\" alt=\"Package version\"></a>\n<a href=\"https://python-poetry.org/\" target=\"_blank\"><img src=\"https://img.shields.io/endpoint?url=https://python-poetry.org/badge/v0.json\" alt=\"Poetry\"></a>\n<a href=\"https://pypi.org/project/ai21\" target=\"_blank\"><img src=\"https://img.shields.io/pypi/pyversions/ai21?color=%2334D058\" alt=\"Supported Python versions\"></a>\n<a href=\"https://github.com/semantic-release/semantic-release\" target=\"_blank\"><img src=\"https://img.shields.io/badge/semantic--release-python-e10079?logo=semantic-release\" alt=\"Semantic Release Support\"></a>\n<a href=\"https://opensource.org/licenses/Apache-2.0\" target=\"_blank\"><img src=\"https://img.shields.io/badge/License-Apache_2.0-blue.svg\" alt=\"License\"></a>\n</p>\n\n---\n\n## Table of Contents\n\n- [Examples](#examples-tldr) 🗂️\n- [AI21 Official Documentation](#Documentation)\n- [Installation](#Installat"},{"ref":"P18","kind":"page","title":"AI21Labs/AI21-Industry-Samples repository metadata","date":"2026-06-11T04:19:59.114182+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/AI21-Industry-Samples","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/AI21-Industry-Samples\n\nDescription: This repository gives sample code for using AI21 Generative AI solutions for industry applications\n\nLanguage: Jupyter Notebook\n\nLicense: Apache-2.0\n\nStars: 18\n\nForks: 8\n\nOpen issues: 1\n\nCreated: 2024-02-02T14:30:39Z\n\nPushed: 2024-12-30T14:55:25Z\n\nDefault branch: main\n\nFork: no\n\nArchived: no\n\nREADME:\n# AI21 Industry Samples\nThis repository gives sample code for using AI21 Generative AI solutions for industry applications. This repository shows how you can use AI21 models in a variety of different ways, including using the AI21 SDK, Amazon SageMaker, and within Amazon Bedrock. The samples also covers a wide variety of use cases and types. \n\nIf you have questions about this repository, feel free to reach out to joshuab@ai21.com"},{"ref":"P19","kind":"page","title":"AI21Labs/salt repository metadata","date":"2026-06-11T04:19:59.111106+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/salt","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/salt\n\nLanguage: Python\n\nLicense: Apache-2.0\n\nStars: 0\n\nForks: 1\n\nOpen issues: 0\n\nCreated: 2024-07-21T09:13:44Z\n\nPushed: 2024-09-23T08:32:19Z\n\nDefault branch: main\n\nFork: no\n\nArchived: yes\n\nREADME:\n## SALT 🧂 Simple Active-Learning Tool\n\nSALT is a Streamlit app for textual data analysis 📚🔬\n\nJust give it a dataset of texts, and it will allow you to:\n\n* Perform semantic-search over the dataset 🔎\n* Label examples efficiently through active-learning 🔄\n* Create a simple & fast (yet surprisingly good) classifier 🤖\n* Find clusters of similar examples (lexically and/or semantically) 🗄️\n\n## How to run?\n1. Clone the repository\n```\ngit clone https://github.com/AI21Labs/salt.git\ncd salt\n```\n\n2. (Optional) Set up virtual environment\n```\npyenv virtualenv 3.9.0 salt\npyenv activate salt\npip install --upgrade pip\n```\n\n3. Install python dependencies\n```\npip install -r requirements.txt\n```\n\n4. Run the app\n```\npython -m streamlit run salt/view/main.py\n```\n\n## How to use?\nHere is the basic flow (for more advanced options see the FAQ section below)\n\n#### Create a project ⚙️️\n1. Load a CSV or JSON-lines file\n2. Select the relevant column and choose a name for the project\n3. Click on \"Create project\" and wait until the creation process is completed\n\n#### (Optional) Find clusters 🗄️\n1. Go to the \"Clusters\" step\n2. Choose similarity type and number of clusters\n3. Click on \"Run clustering\" and wait for the clustering process to finish\n4. Review the results (clustering overview / by cluster)\n\n#### Active-Learning 🔄\n1. Go to the \"Review 📖\" step\n2. Define the classes: provide at least one \"seed example\" for each class\n* You can use \"Search 🔎\" to easily find relevant examples\n* Assign each example with its label by editing the \"label\" column\n3. Go to the \"Labeling 🖊️️\" step\n4. Label examples one by one (chosen by the active-learning classifier)\n5. After labeling 10 examples, you'll start to see an updating graph of predictions change-rate\n* It may help you know when to stop (once each class has converged to some stable mode)\n6. At any point, you can go back to \"Review 📖\" to:\n* Download all labels and predictions\n* View labels/predictions for specific examples\n* Add a new"},{"ref":"P20","kind":"page","title":"ai21labs/Jamba-v0.1 model card","date":"2026-06-11T03:28:09.741009+00:00","date_source":null,"source_url":"https://huggingface.co/ai21labs/Jamba-v0.1/raw/main/README.md","signal_url":null,"signal_json_url":null,"text":"---\nlibrary_name: transformers\nlicense: apache-2.0\ntags:\n- jamba\n- mamba\n- moe\n---\n\nThis is the base version of the Jamba model. We’ve since released a better, instruct-tuned version, [Jamba-1.5-Mini](https://huggingface.co/ai21labs/AI21-Jamba-1.5-Mini). For even greater performance, check out the scaled-up [Jamba-1.5-Large](https://huggingface.co/ai21labs/AI21-Jamba-1.5-Large).\n\n# Model Card for Jamba\n\nJamba is a state-of-the-art, hybrid SSM-Transformer LLM. It delivers throughput gains over traditional Transformer-based models, while outperforming or matching the leading models of its size class on most common benchmarks.\n\nJamba is the first production-scale Mamba implementation, which opens up interesting research and application opportunities. While this initial experimentation shows encouraging gains, we expect these to be further enhanced with future optimizations and explorations.\n\nThis model card is for the base version of Jamba. It’s a pretrained, mixture-of-experts (MoE) generative text model, with 12B active parameters and a total of 52B parameters across all experts. It supports a 256K context length, and can fit up to 140K tokens on a single 80GB GPU.\n\nFor full details of this model please read the [white paper](https://arxiv.org/abs/2403.19887) and the [release blog post](https://www.ai21.com/blog/announcing-jamba).\n\n## Model Details\n\n- **Developed by:** [AI21](https://www.ai21.com)\n- **Model type:** Joint Attention and Mamba (Jamba)\n- **License:** Apache 2.0\n- **Context length:** 256K\n- **Knowledge cutoff date:** March 5, 2024\n\n## Usage\n### Presequities\nIn order to use Jamba, it is recommended you use `transformers` version 4.40.0 or higher (version 4.39.0 or higher is required):\n```bash\npip install transformers>=4.40.0\n```\n\nIn order to run optimized Mamba implementations, you first need to install `mamba-ssm` and `causal-conv1d`:\n```bash\npip install mamba-ssm causal-conv1d>=1.2.0\n```\nYou also have to have the model on a CUDA device.\n\nYou can run the model not using the optimized Mamba kernels, but it is **not** recommended as it will result in significantly lower latencies. In order to do that, you'll need to specify `use_mamba_kernels=False` whe"},{"ref":"P21","kind":"page","title":"ai21labs/Jamba-tiny-random model card","date":"2026-06-11T03:28:09.728314+00:00","date_source":null,"source_url":"https://huggingface.co/ai21labs/Jamba-tiny-random/raw/main/README.md","signal_url":null,"signal_json_url":null,"text":"---\nlicense: apache-2.0\n---\n\nThis is a tiny, dummy version of [Jamba](https://huggingface.co/ai21labs/Jamba-v0.1), used for debugging and experimentation over the Jamba architecture.\n\nIt has 128M parameters (instead of 52B), **and is initialized with random weights and did not undergo any training.**"},{"ref":"P22","kind":"page","title":"AI21Labs/langchain repository metadata","date":"2026-06-11T03:23:00.154387+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/langchain","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/langchain\n\nDescription: ⚡ Building applications with LLMs through composability ⚡\n\nLanguage: Python\n\nLicense: MIT\n\nStars: 1\n\nForks: 2\n\nOpen issues: 0\n\nCreated: 2023-12-24T08:05:20Z\n\nPushed: 2024-02-28T09:42:59Z\n\nDefault branch: master\n\nFork: yes\n\nParent repository: langchain-ai/langchain\n\nArchived: yes\n\nREADME:\n# 🦜️🔗 LangChain\n\n⚡ Build context-aware reasoning applications ⚡\n\n[![Release Notes](https://img.shields.io/github/release/langchain-ai/langchain)](https://github.com/langchain-ai/langchain/releases)\n[![CI](https://github.com/langchain-ai/langchain/actions/workflows/check_diffs.yml/badge.svg)](https://github.com/langchain-ai/langchain/actions/workflows/check_diffs.yml)\n[![Downloads](https://static.pepy.tech/badge/langchain/month)](https://pepy.tech/project/langchain)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)\n[![Twitter](https://img.shields.io/twitter/url/https/twitter.com/langchainai.svg?style=social&label=Follow%20%40LangChainAI)](https://twitter.com/langchainai)\n[![](https://dcbadge.vercel.app/api/server/6adMQxSpJS?compact=true&style=flat)](https://discord.gg/6adMQxSpJS)\n[![Open in Dev Containers](https://img.shields.io/static/v1?label=Dev%20Containers&message=Open&color=blue&logo=visualstudiocode)](https://vscode.dev/redirect?url=vscode://ms-vscode-remote.remote-containers/cloneInVolume?url=https://github.com/langchain-ai/langchain)\n[![Open in GitHub Codespaces](https://github.com/codespaces/badge.svg)](https://codespaces.new/langchain-ai/langchain)\n[![GitHub star chart](https://img.shields.io/github/stars/langchain-ai/langchain?style=social)](https://star-history.com/#langchain-ai/langchain)\n[![Dependency Status](https://img.shields.io/librariesio/github/langchain-ai/langchain)](https://libraries.io/github/langchain-ai/langchain)\n[![Open Issues](https://img.shields.io/github/issues-raw/langchain-ai/langchain)](https://github.com/langchain-ai/langchain/issues)\n\nLooking for the JS/TS library? Check out [LangChain.js](https://github.com/langchain-ai/langchainjs).\n\nTo help you ship LangChain apps to production faster, check out [LangSmith](https://smith.langchain.com). \n[LangSmit"},{"ref":"P23","kind":"page","title":"AI21Labs/terraform-provider-awscc repository metadata","date":"2026-06-11T03:23:00.131686+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/terraform-provider-awscc","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/terraform-provider-awscc\n\nDescription: Terraform AWS Cloud Control provider\n\nLanguage: HCL\n\nLicense: MPL-2.0\n\nStars: 0\n\nForks: 0\n\nOpen issues: 0\n\nCreated: 2023-08-13T08:46:57Z\n\nPushed: 2023-08-14T18:57:04Z\n\nDefault branch: main\n\nFork: yes\n\nParent repository: hashicorp/terraform-provider-awscc\n\nArchived: yes\n\nREADME:\n<a href=\"https://terraform.io\">\n<img src=\".github/terraform_logo.svg\" alt=\"Terraform logo\" title=\"Terraform\" align=\"right\" height=\"50\" />\n</a>\n\n# Terraform AWS Cloud Control Provider\n\n*This provider is currently in technical preview. This means some aspects of its design and implementation are not yet considered stable. We are actively looking for community feedback in order to solidify its form.*\n\n- Announcement: [HashiCorp Blog](https://www.hashicorp.com/blog/announcing-terraform-aws-cloud-control-provider-tech-preview)\n- Terraform Website: [terraform.io](https://terraform.io)\n- Provider Documentation: [Terraform Registry](https://registry.terraform.io/providers/hashicorp/awscc/latest)\n- Forum: [discuss.hashicorp.com](https://discuss.hashicorp.com/c/terraform-providers/tf-awscc/)\n- Tutorial: [learn.hashicorp.com](https://learn.hashicorp.com/tutorials/terraform/aws-cloud-control)\n\nThe Terraform AWS Cloud Control Provider is a plugin for Terraform that allows for the full lifecycle management of AWS resources using the AWS CloudFormation Cloud Control API.\nThis provider is maintained internally by the HashiCorp AWS Provider team.\n\n### AWS Cloud Control API\n\nThe [AWS Cloud Control API](https://docs.aws.amazon.com/cloudcontrolapi) is a lightweight proxy API to discover, provision and manage cloud resources through a simple, uniform and predictable control plane.\nThe AWS Cloud Control API supports **C**reate, **R**ead, **U**pdate, **D**elete and **L**ist (CRUDL) operations on any AWS resource that is registered in the [AWS CloudFormation registry](https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/registry.html).\n\n#### Coverage\n\nAt launch a subset of AWS resources which can be managed by CloudFormation are supported, some services use an older CloudFormation schema and cannot be used with Cloud Control. AWS are updating all of the"},{"ref":"P24","kind":"page","title":"AI21Labs/dotfiles repository metadata","date":"2026-06-11T03:23:00.052977+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/dotfiles","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/dotfiles\n\nDescription: :computer: My dot files shared between machines.\n\nLanguage: Shell\n\nLicense: GPL-3.0\n\nStars: 1\n\nForks: 2\n\nOpen issues: 0\n\nCreated: 2017-11-15T11:03:47Z\n\nPushed: 2018-03-20T07:55:31Z\n\nDefault branch: master\n\nFork: yes\n\nParent repository: MikeMcQuaid/dotfiles\n\nArchived: yes\n\nREADME:\n# Dot Files\nMy dot files shared between machines for ZSH, Bash, macOS, Linux, Cygwin and MSYS.\n\n## Installation\nCustomise the contents of [`gitconfig-user`](https://github.com/MikeMcQuaid/dotfiles/blob/master/gitconfig-user).\nThis is used for user-specific customisations of every other file.\n\nRun [`script/setup`](https://github.com/MikeMcQuaid/dotfiles/blob/master/script/setup)\nafter checkout to symlink (or copy) everything in this directory to your home directory.\n\n## Status\nI'm using these on all my personal machines.\n\n[![Build Status](https://travis-ci.org/MikeMcQuaid/dotfiles.svg?branch=master)](https://travis-ci.org/MikeMcQuaid/dotfiles)\n\n## Contact\n[Mike McQuaid](mailto:mike@mikemcquaid.com)\n\n## License\nThese dot files are licensed under the [GPLv3 License](https://en.wikipedia.org/wiki/GNU_General_Public_License).\nThe full license text is available in [LICENSE.txt](https://github.com/MikeMcQuaid/dotfiles/blob/master/LICENSE.txt)."},{"ref":"P25","kind":"page","title":"AI21Labs/strap repository metadata","date":"2026-06-11T03:22:59.948514+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/strap","signal_url":null,"signal_json_url":null,"text":"# AI21Labs/strap\n\nDescription: :boot: Bootstrap your macOS development system.\n\nLanguage: Shell\n\nLicense: MIT\n\nStars: 0\n\nForks: 0\n\nOpen issues: 0\n\nCreated: 2017-11-13T20:35:59Z\n\nPushed: 2019-02-19T14:10:14Z\n\nDefault branch: master\n\nFork: yes\n\nParent repository: MikeMcQuaid/strap\n\nArchived: yes\n\nREADME:\n# Strap\nA script to bootstrap a minimal macOS development system. This does not assume you're doing Ruby/Rails/web development but installs the minimal set of software every macOS developer will want.\n\n## Motivation\nReplacing [Boxen](https://github.com/boxen/boxen/) in [GitHub](https://github.com/) with a better tool. This post outlines the problems with Boxen and requirements for Strap and other tools used by GitHub: https://mikemcquaid.com/2016/06/15/replacing-boxen/\n\n## Features\n- Disables Java in Safari (for better security)\n- Enables the macOS screensaver password immediately (for better security)\n- Enables the macOS application firewall (for better security)\n- Adds a `Found this computer?` message to the login screen (for machine recovery)\n- Enables full-disk encryption and saves the FileVault Recovery Key to the Desktop (for better security)\n- Installs the Xcode Command Line Tools (for compilers and Unix tools)\n- Agree to the Xcode license (for using compilers without prompts)\n- Installs [Homebrew](http://brew.sh) (for installing command-line software)\n- Installs [Homebrew Bundle](https://github.com/Homebrew/homebrew-bundle) (for `bundler`-like `Brewfile` support)\n- Installs [Homebrew Services](https://github.com/Homebrew/homebrew-services) (for managing Homebrew-installed services)\n- Installs [Homebrew Cask](https://github.com/caskroom/homebrew-cask) (for installing graphical software)\n- Installs the latest macOS software updates (for better security)\n- Installs dotfiles from a user's `https://github.com/username/dotfiles` repository and runs `script/setup` to configure them.\n- Installs software from a user's `Brewfile` in their `https://github.com/username/homebrew-brewfile` repository or `.Brewfile` in their home directory.\n- A simple web application to set Git's name, email and GitHub token (needs authorized on any organisations you wish to access)\n- Id"},{"ref":"P26","kind":"page","title":"AI21Labs/terraform-provider-awscc v0.58.0","date":"2026-06-11T03:21:18.680103+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/terraform-provider-awscc/releases/tag/v0.58.0","signal_url":null,"signal_json_url":null,"text":"# v0.58.0\n\nRepository: AI21Labs/terraform-provider-awscc\n\nTag: v0.58.0\n\nPublished: 2023-08-14T10:55:00Z\n\nPrerelease: no\n\nRelease notes: none published."},{"ref":"P27","kind":"page","title":"AI21Labs/terraform-provider-awscc v0.59.0","date":"2026-06-11T03:21:18.576013+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/terraform-provider-awscc/releases/tag/v0.59.0","signal_url":null,"signal_json_url":null,"text":"# v0.59.0\n\nRepository: AI21Labs/terraform-provider-awscc\n\nTag: v0.59.0\n\nPublished: 2023-08-14T14:15:59Z\n\nPrerelease: no\n\nRelease notes: none published."},{"ref":"P28","kind":"page","title":"AI21Labs/terraform-provider-awscc v0.59.1","date":"2026-06-11T03:21:18.547032+00:00","date_source":null,"source_url":"https://github.com/AI21Labs/terraform-provider-awscc/releases/tag/v0.59.1","signal_url":null,"signal_json_url":null,"text":"# v0.59.1\n\nRepository: AI21Labs/terraform-provider-awscc\n\nTag: v0.59.1\n\nPublished: 2023-08-14T19:19:44Z\n\nPrerelease: no\n\nRelease notes: none published."},{"ref":"E1","kind":"event","title":"ai21labs/Jamba-v0.1","date":"2024-03-28T12:32:22+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/Jamba-v0.1","signal_url":"https://onlylabs.fyi/signals/7a17f631-ad58-4a31-a574-ab8ce86ef27d","signal_json_url":"https://onlylabs.fyi/signals/7a17f631-ad58-4a31-a574-ab8ce86ef27d/signal.json","text":"model_released · ai21labs/Jamba-v0.1 · signal_desk=releases · occurred_at=2024-03-28T12:32:22+00:00 · url=https://huggingface.co/ai21labs/Jamba-v0.1 · hf_downloads=2321 · hf_likes=1192 · hf_params=51570323328 · pipeline=text-generation · license=apache-2.0"},{"ref":"E2","kind":"event","title":"ai21labs/AI21-Jamba-Mini-1.5","date":"2024-08-19T07:04:45+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/AI21-Jamba-Mini-1.5","signal_url":"https://onlylabs.fyi/signals/e87e93b7-907b-4103-b55a-b1575b5fb83e","signal_json_url":"https://onlylabs.fyi/signals/e87e93b7-907b-4103-b55a-b1575b5fb83e/signal.json","text":"model_released · ai21labs/AI21-Jamba-Mini-1.5 · signal_desk=releases · occurred_at=2024-08-19T07:04:45+00:00 · url=https://huggingface.co/ai21labs/AI21-Jamba-Mini-1.5 · hf_downloads=24340 · hf_likes=269 · hf_params=51570323328 · pipeline=text-generation · license=other"},{"ref":"E3","kind":"event","title":"ai21labs/AI21-Jamba-Large-1.5","date":"2024-08-19T07:05:56+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/AI21-Jamba-Large-1.5","signal_url":"https://onlylabs.fyi/signals/e898475d-5104-4943-a62b-3bb9dcbd852e","signal_json_url":"https://onlylabs.fyi/signals/e898475d-5104-4943-a62b-3bb9dcbd852e/signal.json","text":"model_released · ai21labs/AI21-Jamba-Large-1.5 · signal_desk=releases · occurred_at=2024-08-19T07:05:56+00:00 · url=https://huggingface.co/ai21labs/AI21-Jamba-Large-1.5 · hf_downloads=515 · hf_likes=220 · hf_params=398555145696 · pipeline=text-generation · license=other"},{"ref":"E4","kind":"event","title":"ai21labs/AI21-Jamba-Reasoning-3B","date":"2025-10-05T09:43:05+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/AI21-Jamba-Reasoning-3B","signal_url":"https://onlylabs.fyi/signals/08cad506-5c93-478b-b35e-89c47e0add0b","signal_json_url":"https://onlylabs.fyi/signals/08cad506-5c93-478b-b35e-89c47e0add0b/signal.json","text":"model_released · ai21labs/AI21-Jamba-Reasoning-3B · signal_desk=releases · occurred_at=2025-10-05T09:43:05+00:00 · url=https://huggingface.co/ai21labs/AI21-Jamba-Reasoning-3B · hf_downloads=1378 · hf_likes=137 · hf_params=3197109632 · pipeline=text-generation · license=apache-2.0"},{"ref":"E5","kind":"event","title":"ai21labs/Jamba-tiny-dev","date":"2024-09-03T09:12:56+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/Jamba-tiny-dev","signal_url":"https://onlylabs.fyi/signals/45eeae10-836a-4dc0-ad8e-5247ccba1852","signal_json_url":"https://onlylabs.fyi/signals/45eeae10-836a-4dc0-ad8e-5247ccba1852/signal.json","text":"model_released · ai21labs/Jamba-tiny-dev · signal_desk=releases · occurred_at=2024-09-03T09:12:56+00:00 · url=https://huggingface.co/ai21labs/Jamba-tiny-dev · hf_downloads=1180501 · hf_likes=14 · hf_params=318688640 · license=apache-2.0"},{"ref":"E6","kind":"event","title":"ai21labs/AI21-Jamba-Large-1.6","date":"2025-02-27T16:00:42+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/AI21-Jamba-Large-1.6","signal_url":"https://onlylabs.fyi/signals/daf28eea-a467-4895-b12b-c459437bf9b0","signal_json_url":"https://onlylabs.fyi/signals/daf28eea-a467-4895-b12b-c459437bf9b0/signal.json","text":"model_released · ai21labs/AI21-Jamba-Large-1.6 · signal_desk=releases · occurred_at=2025-02-27T16:00:42+00:00 · url=https://huggingface.co/ai21labs/AI21-Jamba-Large-1.6 · hf_downloads=504 · hf_likes=72 · hf_params=398555145696 · pipeline=text-generation · license=other"},{"ref":"E7","kind":"event","title":"ai21labs/AI21-Jamba-Mini-1.6","date":"2025-02-27T16:01:34+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/AI21-Jamba-Mini-1.6","signal_url":"https://onlylabs.fyi/signals/6ec78ce8-1707-45e3-8a8a-2f84fb5bfaf7","signal_json_url":"https://onlylabs.fyi/signals/6ec78ce8-1707-45e3-8a8a-2f84fb5bfaf7/signal.json","text":"model_released · ai21labs/AI21-Jamba-Mini-1.6 · signal_desk=releases · occurred_at=2025-02-27T16:01:34+00:00 · url=https://huggingface.co/ai21labs/AI21-Jamba-Mini-1.6 · hf_downloads=563 · hf_likes=57 · hf_params=51570323328 · pipeline=text-generation · license=other"},{"ref":"E8","kind":"event","title":"ai21labs/AI21-Jamba2-Mini","date":"2026-01-06T08:18:34+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/AI21-Jamba2-Mini","signal_url":"https://onlylabs.fyi/signals/c335fa3f-d9d7-48f5-b630-524d18848151","signal_json_url":"https://onlylabs.fyi/signals/c335fa3f-d9d7-48f5-b630-524d18848151/signal.json","text":"model_released · ai21labs/AI21-Jamba2-Mini · signal_desk=releases · occurred_at=2026-01-06T08:18:34+00:00 · url=https://huggingface.co/ai21labs/AI21-Jamba2-Mini · hf_downloads=673 · hf_likes=54 · hf_params=51570323328 · pipeline=text-generation · license=apache-2.0"},{"ref":"E9","kind":"event","title":"ai21labs/AI21-Jamba2-3B","date":"2026-01-06T08:19:07+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/AI21-Jamba2-3B","signal_url":"https://onlylabs.fyi/signals/82bd31a3-1060-481c-8dca-c67a3460081d","signal_json_url":"https://onlylabs.fyi/signals/82bd31a3-1060-481c-8dca-c67a3460081d/signal.json","text":"model_released · ai21labs/AI21-Jamba2-3B · signal_desk=releases · occurred_at=2026-01-06T08:19:07+00:00 · url=https://huggingface.co/ai21labs/AI21-Jamba2-3B · hf_downloads=560 · hf_likes=42 · pipeline=text-generation · license=apache-2.0"},{"ref":"E10","kind":"event","title":"ai21labs/AI21-Jamba-Mini-1.7","date":"2025-07-01T10:46:27+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/AI21-Jamba-Mini-1.7","signal_url":"https://onlylabs.fyi/signals/f45882cb-77d4-43d1-9894-de46e865960f","signal_json_url":"https://onlylabs.fyi/signals/f45882cb-77d4-43d1-9894-de46e865960f/signal.json","text":"model_released · ai21labs/AI21-Jamba-Mini-1.7 · signal_desk=releases · occurred_at=2025-07-01T10:46:27+00:00 · url=https://huggingface.co/ai21labs/AI21-Jamba-Mini-1.7 · hf_downloads=70 · hf_likes=42 · hf_params=51570323328 · license=other"},{"ref":"E11","kind":"event","title":"ai21labs/AI21-Jamba-Large-1.7","date":"2025-07-02T07:07:10+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/AI21-Jamba-Large-1.7","signal_url":"https://onlylabs.fyi/signals/12cb53f1-8243-4247-837b-ee88d2015ed7","signal_json_url":"https://onlylabs.fyi/signals/12cb53f1-8243-4247-837b-ee88d2015ed7/signal.json","text":"model_released · ai21labs/AI21-Jamba-Large-1.7 · signal_desk=releases · occurred_at=2025-07-02T07:07:10+00:00 · url=https://huggingface.co/ai21labs/AI21-Jamba-Large-1.7 · hf_downloads=29 · hf_likes=34 · hf_params=398555145696 · license=other"},{"ref":"E12","kind":"event","title":"AI21Labs/pre-commit-hadolint","date":"2026-05-12T08:34:02+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/pre-commit-hadolint","signal_url":"https://onlylabs.fyi/signals/a8d5a050-6c2c-48fc-8704-64a597e69658","signal_json_url":"https://onlylabs.fyi/signals/a8d5a050-6c2c-48fc-8704-64a597e69658/signal.json","text":"repo_new · AI21Labs/pre-commit-hadolint · signal_desk=repos · occurred_at=2026-05-12T08:34:02+00:00 · url=https://github.com/AI21Labs/pre-commit-hadolint · stars=3 · raw={\"repo\":\"AI21Labs/pre-commit-hadolint\",\"description\":\"Pre-commit hooks for Dockerfile linter\",\"language\":\"Python\"}"},{"ref":"E13","kind":"event","title":"ai21labs/Jamba-tiny-random","date":"2024-04-17T00:04:25+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/Jamba-tiny-random","signal_url":"https://onlylabs.fyi/signals/f145d8e0-b0e1-4719-a69a-7e89ed01cf80","signal_json_url":"https://onlylabs.fyi/signals/f145d8e0-b0e1-4719-a69a-7e89ed01cf80/signal.json","text":"model_released · ai21labs/Jamba-tiny-random · signal_desk=releases · occurred_at=2024-04-17T00:04:25+00:00 · url=https://huggingface.co/ai21labs/Jamba-tiny-random · hf_downloads=10736 · hf_likes=13 · hf_params=127679344 · pipeline=text-generation · license=apache-2.0"},{"ref":"E14","kind":"event","title":"ai21labs/Jamba-tiny-reward-dev","date":"2024-12-05T14:57:37+00:00","date_source":"source","source_url":"https://huggingface.co/ai21labs/Jamba-tiny-reward-dev","signal_url":"https://onlylabs.fyi/signals/f4d67d57-7db3-4e7e-a11e-1c78c79a292c","signal_json_url":"https://onlylabs.fyi/signals/f4d67d57-7db3-4e7e-a11e-1c78c79a292c/signal.json","text":"model_released · ai21labs/Jamba-tiny-reward-dev · signal_desk=releases · occurred_at=2024-12-05T14:57:37+00:00 · url=https://huggingface.co/ai21labs/Jamba-tiny-reward-dev · hf_downloads=21729 · hf_likes=2 · license=apache-2.0"},{"ref":"E15","kind":"event","title":"AI21Labs/in-context-ralm","date":"2023-01-26T12:07:18+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/in-context-ralm","signal_url":"https://onlylabs.fyi/signals/db1fb1e2-5bfa-4704-8c89-f77fac3c5cd1","signal_json_url":"https://onlylabs.fyi/signals/db1fb1e2-5bfa-4704-8c89-f77fac3c5cd1/signal.json","text":"repo_new · AI21Labs/in-context-ralm · signal_desk=repos · occurred_at=2023-01-26T12:07:18+00:00 · url=https://github.com/AI21Labs/in-context-ralm · stars=295 · raw={\"repo\":\"AI21Labs/in-context-ralm\",\"language\":\"Python\"}"},{"ref":"E16","kind":"event","title":"AI21Labs/lm-evaluation","date":"2021-08-05T13:22:20+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/lm-evaluation","signal_url":"https://onlylabs.fyi/signals/69e15875-922b-4d38-a2be-4015751f183a","signal_json_url":"https://onlylabs.fyi/signals/69e15875-922b-4d38-a2be-4015751f183a/signal.json","text":"repo_new · AI21Labs/lm-evaluation · signal_desk=repos · occurred_at=2021-08-05T13:22:20+00:00 · url=https://github.com/AI21Labs/lm-evaluation · stars=130 · raw={\"repo\":\"AI21Labs/lm-evaluation\",\"description\":\"Evaluation suite for large-scale language models.\",\"language\":\"Python\"}"},{"ref":"E17","kind":"event","title":"AI21Labs/Parallel-Context-Windows","date":"2023-02-22T13:14:24+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/Parallel-Context-Windows","signal_url":"https://onlylabs.fyi/signals/68e21639-b576-444f-9b9d-6971f0b6739e","signal_json_url":"https://onlylabs.fyi/signals/68e21639-b576-444f-9b9d-6971f0b6739e/signal.json","text":"repo_new · AI21Labs/Parallel-Context-Windows · signal_desk=repos · occurred_at=2023-02-22T13:14:24+00:00 · url=https://github.com/AI21Labs/Parallel-Context-Windows · stars=107 · raw={\"repo\":\"AI21Labs/Parallel-Context-Windows\",\"language\":\"Python\"}"},{"ref":"E18","kind":"event","title":"AI21Labs/ai21-python","date":"2023-12-10T09:20:15+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-python","signal_url":"https://onlylabs.fyi/signals/3ceb6253-41fc-44c7-a33e-62f570e3922b","signal_json_url":"https://onlylabs.fyi/signals/3ceb6253-41fc-44c7-a33e-62f570e3922b/signal.json","text":"repo_new · AI21Labs/ai21-python · signal_desk=repos · occurred_at=2023-12-10T09:20:15+00:00 · url=https://github.com/AI21Labs/ai21-python · stars=70 · raw={\"repo\":\"AI21Labs/ai21-python\",\"description\":\"AI21 Python SDK\",\"language\":\"Python\"}"},{"ref":"E19","kind":"event","title":"AI21Labs/factor","date":"2023-07-13T12:15:34+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/factor","signal_url":"https://onlylabs.fyi/signals/0951a0ce-c4b5-4fee-bef1-30836ae14246","signal_json_url":"https://onlylabs.fyi/signals/0951a0ce-c4b5-4fee-bef1-30836ae14246/signal.json","text":"repo_new · AI21Labs/factor · signal_desk=repos · occurred_at=2023-07-13T12:15:34+00:00 · url=https://github.com/AI21Labs/factor · stars=53 · raw={\"repo\":\"AI21Labs/factor\",\"description\":\"Code and data for the FACTOR paper\",\"language\":\"Python\"}"},{"ref":"E20","kind":"event","title":"AI21Labs/sense-bert","date":"2020-07-06T07:54:34+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/sense-bert","signal_url":"https://onlylabs.fyi/signals/74700193-da5c-4b6d-92d8-781a0b8941fc","signal_json_url":"https://onlylabs.fyi/signals/74700193-da5c-4b6d-92d8-781a0b8941fc/signal.json","text":"repo_new · AI21Labs/sense-bert · signal_desk=repos · occurred_at=2020-07-06T07:54:34+00:00 · url=https://github.com/AI21Labs/sense-bert · stars=48 · raw={\"repo\":\"AI21Labs/sense-bert\",\"description\":\"This is the code for loading the SenseBERT model, described in our paper from ACL 2020.\",\"language\":\"Python\"}"},{"ref":"E21","kind":"event","title":"AI21Labs/ai21-tokenizer","date":"2023-08-22T08:28:46+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-tokenizer","signal_url":"https://onlylabs.fyi/signals/3654e90c-8a76-4a9f-b80e-e7e89c4edb52","signal_json_url":"https://onlylabs.fyi/signals/3654e90c-8a76-4a9f-b80e-e7e89c4edb52/signal.json","text":"repo_new · AI21Labs/ai21-tokenizer · signal_desk=repos · occurred_at=2023-08-22T08:28:46+00:00 · url=https://github.com/AI21Labs/ai21-tokenizer · stars=33 · raw={\"repo\":\"AI21Labs/ai21-tokenizer\",\"description\":\"AI21's Jamba models tokenizers\",\"language\":\"Python\"}"},{"ref":"E22","kind":"event","title":"AI21Labs/SageMaker","date":"2022-11-22T14:42:37+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/SageMaker","signal_url":"https://onlylabs.fyi/signals/e5672c3c-e39b-4ef5-96a5-b444994e393d","signal_json_url":"https://onlylabs.fyi/signals/e5672c3c-e39b-4ef5-96a5-b444994e393d/signal.json","text":"repo_new · AI21Labs/SageMaker · signal_desk=repos · occurred_at=2022-11-22T14:42:37+00:00 · url=https://github.com/AI21Labs/SageMaker · stars=32 · raw={\"repo\":\"AI21Labs/SageMaker\",\"description\":\"Examples for using AI21's models through Amazon SageMaker\",\"language\":\"Jupyter Notebook\"}"},{"ref":"E23","kind":"event","title":"AI21Labs/AI21-Industry-Samples","date":"2024-02-02T14:30:39+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/AI21-Industry-Samples","signal_url":"https://onlylabs.fyi/signals/d4083267-defa-4b38-8f7f-90ee756c2801","signal_json_url":"https://onlylabs.fyi/signals/d4083267-defa-4b38-8f7f-90ee756c2801/signal.json","text":"repo_new · AI21Labs/AI21-Industry-Samples · signal_desk=repos · occurred_at=2024-02-02T14:30:39+00:00 · url=https://github.com/AI21Labs/AI21-Industry-Samples · stars=18 · raw={\"repo\":\"AI21Labs/AI21-Industry-Samples\",\"description\":\"This repository gives sample code for using AI21 Generative AI solutions for industry applications\",\"language\":\"Jupyter Notebook\"}"},{"ref":"E24","kind":"event","title":"AI21Labs/Rick-and-Morty-episode-generator","date":"2023-01-12T10:32:39+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/Rick-and-Morty-episode-generator","signal_url":"https://onlylabs.fyi/signals/83cdac26-60ba-43dc-a855-36327bf1c1b5","signal_json_url":"https://onlylabs.fyi/signals/83cdac26-60ba-43dc-a855-36327bf1c1b5/signal.json","text":"repo_new · AI21Labs/Rick-and-Morty-episode-generator · signal_desk=repos · occurred_at=2023-01-12T10:32:39+00:00 · url=https://github.com/AI21Labs/Rick-and-Morty-episode-generator · stars=16 · raw={\"repo\":\"AI21Labs/Rick-and-Morty-episode-generator\",\"description\":\"An interactive one-pager browser game, primarily built with React.js. It makes use of the state-of-the-art Jurassic-2 language models to generate made-up episodes of the popular animated science fiction TV show Rick and Morty.\",\"language\":\"JavaScript\"}"},{"ref":"E25","kind":"event","title":"AI21Labs/pmi-masking","date":"2021-07-07T13:35:35+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/pmi-masking","signal_url":"https://onlylabs.fyi/signals/7cb6383a-3088-4f3d-a2d2-efb7843da1d5","signal_json_url":"https://onlylabs.fyi/signals/7cb6383a-3088-4f3d-a2d2-efb7843da1d5/signal.json","text":"repo_new · AI21Labs/pmi-masking · signal_desk=repos · occurred_at=2021-07-07T13:35:35+00:00 · url=https://github.com/AI21Labs/pmi-masking · stars=14 · raw={\"repo\":\"AI21Labs/pmi-masking\",\"description\":\"This repository includes the masking vocabulary used in the ICLR 2021 spotlight PMI-Masking paper\"}"},{"ref":"E26","kind":"event","title":"AI21Labs/ai21-typescript","date":"2024-11-05T09:44:25+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-typescript","signal_url":"https://onlylabs.fyi/signals/ac9e3f80-ea99-4dba-8929-da82435a3a34","signal_json_url":"https://onlylabs.fyi/signals/ac9e3f80-ea99-4dba-8929-da82435a3a34/signal.json","text":"repo_new · AI21Labs/ai21-typescript · signal_desk=repos · occurred_at=2024-11-05T09:44:25+00:00 · url=https://github.com/AI21Labs/ai21-typescript · stars=13 · raw={\"repo\":\"AI21Labs/ai21-typescript\",\"description\":\"AI21 Typescript SDK\",\"language\":\"TypeScript\"}"},{"ref":"E27","kind":"event","title":"AI21Labs/multi-window-chunk-size","date":"2026-01-26T12:02:59+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/multi-window-chunk-size","signal_url":"https://onlylabs.fyi/signals/7beb24f5-17a5-4573-92ef-792037a0120f","signal_json_url":"https://onlylabs.fyi/signals/7beb24f5-17a5-4573-92ef-792037a0120f/signal.json","text":"repo_new · AI21Labs/multi-window-chunk-size · signal_desk=repos · occurred_at=2026-01-26T12:02:59+00:00 · url=https://github.com/AI21Labs/multi-window-chunk-size · stars=7 · raw={\"repo\":\"AI21Labs/multi-window-chunk-size\",\"language\":\"Jupyter Notebook\"}"},{"ref":"E28","kind":"event","title":"AI21Labs/MRKL_synthetic_data","date":"2022-04-27T06:51:01+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/MRKL_synthetic_data","signal_url":"https://onlylabs.fyi/signals/36893f34-fa22-40b2-b363-3773f1f43059","signal_json_url":"https://onlylabs.fyi/signals/36893f34-fa22-40b2-b363-3773f1f43059/signal.json","text":"repo_new · AI21Labs/MRKL_synthetic_data · signal_desk=repos · occurred_at=2022-04-27T06:51:01+00:00 · url=https://github.com/AI21Labs/MRKL_synthetic_data · stars=5 · raw={\"repo\":\"AI21Labs/MRKL_synthetic_data\",\"description\":\"All synthetic data used for the experiments described in MRKL white paper https://www.ai21.com/blog/jurassic-x-crossing-the-neuro-symbolic-chasm-with-the-mrkl-system.\"}"},{"ref":"E29","kind":"event","title":"AI21Labs/jurassic-chess","date":"2021-08-10T06:51:54+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/jurassic-chess","signal_url":"https://onlylabs.fyi/signals/05c706b6-d4b4-4c58-85f2-cf5315a2ca5c","signal_json_url":"https://onlylabs.fyi/signals/05c706b6-d4b4-4c58-85f2-cf5315a2ca5c/signal.json","text":"repo_new · AI21Labs/jurassic-chess · signal_desk=repos · occurred_at=2021-08-10T06:51:54+00:00 · url=https://github.com/AI21Labs/jurassic-chess · stars=5 · raw={\"repo\":\"AI21Labs/jurassic-chess\",\"description\":\"A talking chess AI\",\"language\":\"Python\"}"},{"ref":"E30","kind":"event","title":"AI21Labs/jurades","date":"2021-08-04T18:16:41+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/jurades","signal_url":"https://onlylabs.fyi/signals/e08d99bc-4272-4cda-9912-494ac03a1a25","signal_json_url":"https://onlylabs.fyi/signals/e08d99bc-4272-4cda-9912-494ac03a1a25/signal.json","text":"repo_new · AI21Labs/jurades · signal_desk=repos · occurred_at=2021-08-04T18:16:41+00:00 · url=https://github.com/AI21Labs/jurades · stars=3 · raw={\"repo\":\"AI21Labs/jurades\",\"description\":\"AI Charades game\",\"language\":\"JavaScript\"}"},{"ref":"E31","kind":"event","title":"AI21Labs/transformers","date":"2024-03-28T14:42:27+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/transformers","signal_url":"https://onlylabs.fyi/signals/2a66d88d-76ee-4b29-b828-bb748c12875f","signal_json_url":"https://onlylabs.fyi/signals/2a66d88d-76ee-4b29-b828-bb748c12875f/signal.json","text":"repo_forked · AI21Labs/transformers · signal_desk=forks · occurred_at=2024-03-28T14:42:27+00:00 · url=https://github.com/AI21Labs/transformers · stars=1 · raw={\"repo\":\"AI21Labs/transformers\",\"parent\":\"huggingface/transformers\"}"},{"ref":"E32","kind":"event","title":"AI21Labs/langchain","date":"2023-12-24T08:05:20+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/langchain","signal_url":"https://onlylabs.fyi/signals/a859f4f1-62ca-4332-89ec-2ce11c22aafb","signal_json_url":"https://onlylabs.fyi/signals/a859f4f1-62ca-4332-89ec-2ce11c22aafb/signal.json","text":"repo_forked · AI21Labs/langchain · signal_desk=forks · occurred_at=2023-12-24T08:05:20+00:00 · url=https://github.com/AI21Labs/langchain · stars=1 · raw={\"repo\":\"AI21Labs/langchain\",\"parent\":\"langchain-ai/langchain\"}"},{"ref":"E33","kind":"event","title":"AI21Labs/dev-envs","date":"2022-12-20T11:48:51+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/dev-envs","signal_url":"https://onlylabs.fyi/signals/214d8ed3-9c06-46f5-a86b-9809a718ca49","signal_json_url":"https://onlylabs.fyi/signals/214d8ed3-9c06-46f5-a86b-9809a718ca49/signal.json","text":"repo_new · AI21Labs/dev-envs · signal_desk=repos · occurred_at=2022-12-20T11:48:51+00:00 · url=https://github.com/AI21Labs/dev-envs · stars=1 · raw={\"repo\":\"AI21Labs/dev-envs\",\"language\":\"Shell\"}"},{"ref":"E34","kind":"event","title":"AI21Labs/github-migration","date":"2022-12-20T10:55:34+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/github-migration","signal_url":"https://onlylabs.fyi/signals/497344a4-a27b-49fe-b0f8-5ff1a174d3ba","signal_json_url":"https://onlylabs.fyi/signals/497344a4-a27b-49fe-b0f8-5ff1a174d3ba/signal.json","text":"repo_new · AI21Labs/github-migration · signal_desk=repos · occurred_at=2022-12-20T10:55:34+00:00 · url=https://github.com/AI21Labs/github-migration · stars=1 · raw={\"repo\":\"AI21Labs/github-migration\",\"language\":\"JavaScript\"}"},{"ref":"E35","kind":"event","title":"AI21Labs/dotfiles","date":"2017-11-15T11:03:47+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/dotfiles","signal_url":"https://onlylabs.fyi/signals/c551bd23-8954-4b01-a855-a9887e46c9f4","signal_json_url":"https://onlylabs.fyi/signals/c551bd23-8954-4b01-a855-a9887e46c9f4/signal.json","text":"repo_forked · AI21Labs/dotfiles · signal_desk=forks · occurred_at=2017-11-15T11:03:47+00:00 · url=https://github.com/AI21Labs/dotfiles · stars=1 · raw={\"repo\":\"AI21Labs/dotfiles\",\"parent\":\"MikeMcQuaid/dotfiles\"}"},{"ref":"E36","kind":"event","title":"AI21Labs/auto-tuning-vllm","date":"2025-12-24T09:58:31+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/auto-tuning-vllm","signal_url":"https://onlylabs.fyi/signals/157e2977-d907-4dd3-bafb-79a255ee690c","signal_json_url":"https://onlylabs.fyi/signals/157e2977-d907-4dd3-bafb-79a255ee690c/signal.json","text":"repo_forked · AI21Labs/auto-tuning-vllm · signal_desk=forks · occurred_at=2025-12-24T09:58:31+00:00 · url=https://github.com/AI21Labs/auto-tuning-vllm · raw={\"repo\":\"AI21Labs/auto-tuning-vllm\",\"parent\":\"openshift-psap/auto-tuning-vllm\"}"},{"ref":"E37","kind":"event","title":"AI21Labs/ai21-python v4.3.0","date":"2025-11-06T13:56:23+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-python/releases/tag/v4.3.0","signal_url":"https://onlylabs.fyi/signals/13bf3143-e623-4aa1-a76a-2d6f3ff1a552","signal_json_url":"https://onlylabs.fyi/signals/13bf3143-e623-4aa1-a76a-2d6f3ff1a552/signal.json","text":"release · AI21Labs/ai21-python v4.3.0 · signal_desk=releases · occurred_at=2025-11-06T13:56:23+00:00 · url=https://github.com/AI21Labs/ai21-python/releases/tag/v4.3.0 · raw={\"repo\":\"AI21Labs/ai21-python\"}"},{"ref":"E38","kind":"event","title":"AI21Labs/ai21-python v4.2.1","date":"2025-10-28T17:35:30+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-python/releases/tag/v4.2.1","signal_url":"https://onlylabs.fyi/signals/718a8ddb-49e1-4cbf-aae9-0ed19e73d703","signal_json_url":"https://onlylabs.fyi/signals/718a8ddb-49e1-4cbf-aae9-0ed19e73d703/signal.json","text":"release · AI21Labs/ai21-python v4.2.1 · signal_desk=releases · occurred_at=2025-10-28T17:35:30+00:00 · url=https://github.com/AI21Labs/ai21-python/releases/tag/v4.2.1 · raw={\"repo\":\"AI21Labs/ai21-python\"}"},{"ref":"E39","kind":"event","title":"AI21Labs/ai21-typescript v1.3.0","date":"2025-09-20T17:33:39+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.3.0","signal_url":"https://onlylabs.fyi/signals/78e89add-5a2e-44c5-bf32-51af7846a4c6","signal_json_url":"https://onlylabs.fyi/signals/78e89add-5a2e-44c5-bf32-51af7846a4c6/signal.json","text":"release · AI21Labs/ai21-typescript v1.3.0 · signal_desk=releases · occurred_at=2025-09-20T17:33:39+00:00 · url=https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.3.0 · raw={\"repo\":\"AI21Labs/ai21-typescript\"}"},{"ref":"E40","kind":"event","title":"AI21Labs/ai21-typescript v1.2.0","date":"2025-09-18T14:40:00+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.2.0","signal_url":"https://onlylabs.fyi/signals/52ca6d4f-01b4-465d-b052-99d26ce06028","signal_json_url":"https://onlylabs.fyi/signals/52ca6d4f-01b4-465d-b052-99d26ce06028/signal.json","text":"release · AI21Labs/ai21-typescript v1.2.0 · signal_desk=releases · occurred_at=2025-09-18T14:40:00+00:00 · url=https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.2.0 · raw={\"repo\":\"AI21Labs/ai21-typescript\"}"},{"ref":"E41","kind":"event","title":"AI21Labs/ai21-python v4.2.0","date":"2025-09-18T14:19:58+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-python/releases/tag/v4.2.0","signal_url":"https://onlylabs.fyi/signals/291df13e-eceb-4515-9426-04c25bb584ee","signal_json_url":"https://onlylabs.fyi/signals/291df13e-eceb-4515-9426-04c25bb584ee/signal.json","text":"release · AI21Labs/ai21-python v4.2.0 · signal_desk=releases · occurred_at=2025-09-18T14:19:58+00:00 · url=https://github.com/AI21Labs/ai21-python/releases/tag/v4.2.0 · raw={\"repo\":\"AI21Labs/ai21-python\"}"},{"ref":"E42","kind":"event","title":"AI21Labs/ai21-python v4.1.1","date":"2025-09-16T07:20:40+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-python/releases/tag/v4.1.1","signal_url":"https://onlylabs.fyi/signals/c1e0a531-0139-4e3e-a5b5-c4141f0dcd64","signal_json_url":"https://onlylabs.fyi/signals/c1e0a531-0139-4e3e-a5b5-c4141f0dcd64/signal.json","text":"release · AI21Labs/ai21-python v4.1.1 · signal_desk=releases · occurred_at=2025-09-16T07:20:40+00:00 · url=https://github.com/AI21Labs/ai21-python/releases/tag/v4.1.1 · raw={\"repo\":\"AI21Labs/ai21-python\"}"},{"ref":"E43","kind":"event","title":"AI21Labs/ai21-python v4.1.0","date":"2025-08-19T06:54:53+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-python/releases/tag/v4.1.0","signal_url":"https://onlylabs.fyi/signals/ddf75f8f-adc7-49ef-b4e9-aa870a91d06d","signal_json_url":"https://onlylabs.fyi/signals/ddf75f8f-adc7-49ef-b4e9-aa870a91d06d/signal.json","text":"release · AI21Labs/ai21-python v4.1.0 · signal_desk=releases · occurred_at=2025-08-19T06:54:53+00:00 · url=https://github.com/AI21Labs/ai21-python/releases/tag/v4.1.0 · raw={\"repo\":\"AI21Labs/ai21-python\"}"},{"ref":"E44","kind":"event","title":"AI21Labs/ai21-tokenizer v1.1.0","date":"2025-06-12T07:08:52+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v1.1.0","signal_url":"https://onlylabs.fyi/signals/8bab774d-b52d-4bd4-8262-c3cc369da9d3","signal_json_url":"https://onlylabs.fyi/signals/8bab774d-b52d-4bd4-8262-c3cc369da9d3/signal.json","text":"release · AI21Labs/ai21-tokenizer v1.1.0 · signal_desk=releases · occurred_at=2025-06-12T07:08:52+00:00 · url=https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v1.1.0 · raw={\"repo\":\"AI21Labs/ai21-tokenizer\"}"},{"ref":"E45","kind":"event","title":"AI21Labs/ai21-tokenizer v1.0.0","date":"2025-05-28T14:35:49+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v1.0.0","signal_url":"https://onlylabs.fyi/signals/6ab0d931-6205-4f32-90d7-5c18ed834908","signal_json_url":"https://onlylabs.fyi/signals/6ab0d931-6205-4f32-90d7-5c18ed834908/signal.json","text":"release · AI21Labs/ai21-tokenizer v1.0.0 · signal_desk=releases · occurred_at=2025-05-28T14:35:49+00:00 · url=https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v1.0.0 · raw={\"repo\":\"AI21Labs/ai21-tokenizer\"}"},{"ref":"E46","kind":"event","title":"AI21Labs/ai21-typescript v1.1.1","date":"2025-03-30T09:14:04+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.1.1","signal_url":"https://onlylabs.fyi/signals/73ac1979-83ee-468d-9b1f-8dc455891cbf","signal_json_url":"https://onlylabs.fyi/signals/73ac1979-83ee-468d-9b1f-8dc455891cbf/signal.json","text":"release · AI21Labs/ai21-typescript v1.1.1 · signal_desk=releases · occurred_at=2025-03-30T09:14:04+00:00 · url=https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.1.1 · raw={\"repo\":\"AI21Labs/ai21-typescript\"}"},{"ref":"E47","kind":"event","title":"AI21Labs/ai21-typescript v1.1.0","date":"2024-12-05T14:30:11+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.1.0","signal_url":"https://onlylabs.fyi/signals/5ec0687e-c5d5-4d49-87ac-814f9fd04af2","signal_json_url":"https://onlylabs.fyi/signals/5ec0687e-c5d5-4d49-87ac-814f9fd04af2/signal.json","text":"release · AI21Labs/ai21-typescript v1.1.0 · signal_desk=releases · occurred_at=2024-12-05T14:30:11+00:00 · url=https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.1.0 · raw={\"repo\":\"AI21Labs/ai21-typescript\"}"},{"ref":"E48","kind":"event","title":"AI21Labs/ai21-typescript v1.1.0-rc.1","date":"2024-12-05T09:59:28+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.1.0-rc.1","signal_url":"https://onlylabs.fyi/signals/2485100e-297d-451f-9652-500dc0c14e30","signal_json_url":"https://onlylabs.fyi/signals/2485100e-297d-451f-9652-500dc0c14e30/signal.json","text":"release · AI21Labs/ai21-typescript v1.1.0-rc.1 · signal_desk=releases · occurred_at=2024-12-05T09:59:28+00:00 · url=https://github.com/AI21Labs/ai21-typescript/releases/tag/v1.1.0-rc.1 · raw={\"repo\":\"AI21Labs/ai21-typescript\"}"},{"ref":"E49","kind":"event","title":"AI21Labs/ai21-tokenizer v0.12.0","date":"2024-08-21T11:41:27+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v0.12.0","signal_url":"https://onlylabs.fyi/signals/f9a90f06-b8bd-469f-9408-3b56fefe1620","signal_json_url":"https://onlylabs.fyi/signals/f9a90f06-b8bd-469f-9408-3b56fefe1620/signal.json","text":"release · AI21Labs/ai21-tokenizer v0.12.0 · signal_desk=releases · occurred_at=2024-08-21T11:41:27+00:00 · url=https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v0.12.0 · raw={\"repo\":\"AI21Labs/ai21-tokenizer\"}"},{"ref":"E50","kind":"event","title":"AI21Labs/ai21-tokenizer v0.11.3","date":"2024-08-05T13:45:16+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v0.11.3","signal_url":"https://onlylabs.fyi/signals/650473de-b22b-42dc-874f-7c5808c6c420","signal_json_url":"https://onlylabs.fyi/signals/650473de-b22b-42dc-874f-7c5808c6c420/signal.json","text":"release · AI21Labs/ai21-tokenizer v0.11.3 · signal_desk=releases · occurred_at=2024-08-05T13:45:16+00:00 · url=https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v0.11.3 · raw={\"repo\":\"AI21Labs/ai21-tokenizer\"}"},{"ref":"E51","kind":"event","title":"AI21Labs/salt","date":"2024-07-21T09:13:44+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/salt","signal_url":"https://onlylabs.fyi/signals/653c8c77-1a4f-4e50-8504-f5f596fb5be3","signal_json_url":"https://onlylabs.fyi/signals/653c8c77-1a4f-4e50-8504-f5f596fb5be3/signal.json","text":"repo_new · AI21Labs/salt · signal_desk=repos · occurred_at=2024-07-21T09:13:44+00:00 · url=https://github.com/AI21Labs/salt · raw={\"repo\":\"AI21Labs/salt\",\"language\":\"Python\"}"},{"ref":"E52","kind":"event","title":"AI21Labs/ai21-tokenizer v0.11.2","date":"2024-06-20T11:30:43+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v0.11.2","signal_url":"https://onlylabs.fyi/signals/b9e907e5-6464-4be1-b333-e073f3d9e26f","signal_json_url":"https://onlylabs.fyi/signals/b9e907e5-6464-4be1-b333-e073f3d9e26f/signal.json","text":"release · AI21Labs/ai21-tokenizer v0.11.2 · signal_desk=releases · occurred_at=2024-06-20T11:30:43+00:00 · url=https://github.com/AI21Labs/ai21-tokenizer/releases/tag/v0.11.2 · raw={\"repo\":\"AI21Labs/ai21-tokenizer\"}"},{"ref":"E53","kind":"event","title":"AI21Labs/terraform-provider-awscc v0.59.1","date":"2023-08-14T19:19:44+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/terraform-provider-awscc/releases/tag/v0.59.1","signal_url":"https://onlylabs.fyi/signals/e5115fcd-ec32-4b44-bca0-79dd90889ea4","signal_json_url":"https://onlylabs.fyi/signals/e5115fcd-ec32-4b44-bca0-79dd90889ea4/signal.json","text":"release · AI21Labs/terraform-provider-awscc v0.59.1 · signal_desk=releases · occurred_at=2023-08-14T19:19:44+00:00 · url=https://github.com/AI21Labs/terraform-provider-awscc/releases/tag/v0.59.1 · raw={\"repo\":\"AI21Labs/terraform-provider-awscc\"}"},{"ref":"E54","kind":"event","title":"AI21Labs/terraform-provider-awscc v0.59.0","date":"2023-08-14T14:15:59+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/terraform-provider-awscc/releases/tag/v0.59.0","signal_url":"https://onlylabs.fyi/signals/81dae748-e8e3-45ca-b298-a3675c203529","signal_json_url":"https://onlylabs.fyi/signals/81dae748-e8e3-45ca-b298-a3675c203529/signal.json","text":"release · AI21Labs/terraform-provider-awscc v0.59.0 · signal_desk=releases · occurred_at=2023-08-14T14:15:59+00:00 · url=https://github.com/AI21Labs/terraform-provider-awscc/releases/tag/v0.59.0 · raw={\"repo\":\"AI21Labs/terraform-provider-awscc\"}"},{"ref":"E55","kind":"event","title":"AI21Labs/terraform-provider-awscc v0.58.0","date":"2023-08-14T10:55:00+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/terraform-provider-awscc/releases/tag/v0.58.0","signal_url":"https://onlylabs.fyi/signals/ecb9779c-d2d0-470b-bdec-1b0dd951ce8a","signal_json_url":"https://onlylabs.fyi/signals/ecb9779c-d2d0-470b-bdec-1b0dd951ce8a/signal.json","text":"release · AI21Labs/terraform-provider-awscc v0.58.0 · signal_desk=releases · occurred_at=2023-08-14T10:55:00+00:00 · url=https://github.com/AI21Labs/terraform-provider-awscc/releases/tag/v0.58.0 · raw={\"repo\":\"AI21Labs/terraform-provider-awscc\"}"},{"ref":"E56","kind":"event","title":"AI21Labs/terraform-provider-awscc","date":"2023-08-13T08:46:57+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/terraform-provider-awscc","signal_url":"https://onlylabs.fyi/signals/300e00f4-63c0-4431-800e-ab3fcc229890","signal_json_url":"https://onlylabs.fyi/signals/300e00f4-63c0-4431-800e-ab3fcc229890/signal.json","text":"repo_forked · AI21Labs/terraform-provider-awscc · signal_desk=forks · occurred_at=2023-08-13T08:46:57+00:00 · url=https://github.com/AI21Labs/terraform-provider-awscc · raw={\"repo\":\"AI21Labs/terraform-provider-awscc\",\"parent\":\"hashicorp/terraform-provider-awscc\"}"},{"ref":"E57","kind":"event","title":"AI21Labs/corefact","date":"2023-07-10T12:35:36+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/corefact","signal_url":"https://onlylabs.fyi/signals/0277f707-ab04-44b3-bf1a-9365bf5c4c41","signal_json_url":"https://onlylabs.fyi/signals/0277f707-ab04-44b3-bf1a-9365bf5c4c41/signal.json","text":"repo_new · AI21Labs/corefact · signal_desk=repos · occurred_at=2023-07-10T12:35:36+00:00 · url=https://github.com/AI21Labs/corefact · raw={\"repo\":\"AI21Labs/corefact\"}"},{"ref":"E58","kind":"event","title":"AI21Labs/strap","date":"2017-11-13T20:35:59+00:00","date_source":"source","source_url":"https://github.com/AI21Labs/strap","signal_url":"https://onlylabs.fyi/signals/2be5f5d7-db1a-4dd6-8b8f-03f2811508b9","signal_json_url":"https://onlylabs.fyi/signals/2be5f5d7-db1a-4dd6-8b8f-03f2811508b9/signal.json","text":"repo_forked · AI21Labs/strap · signal_desk=forks · occurred_at=2017-11-13T20:35:59+00:00 · url=https://github.com/AI21Labs/strap · raw={\"repo\":\"AI21Labs/strap\",\"parent\":\"MikeMcQuaid/strap\"}"}]}