{"schema_version":"onlylabs.public_signal.v1","title":"DeepInfra Fork: deepinfra/flash-attention","description":"DeepInfra fork signal with public source context, captured evidence pages, related signals, and category-scoped analysis context.","url":"https://onlylabs.fyi/signals/2a903af5-9d24-4c22-b727-ad90821882dd","json_url":"https://onlylabs.fyi/signals/2a903af5-9d24-4c22-b727-ad90821882dd/signal.json","generated_at":"2026-06-11T02:54:23.295206+00:00","org":{"slug":"deepinfra","name":"DeepInfra","category":"neocloud","category_label":"Neocloud","dossier_url":"https://onlylabs.fyi/labs/deepinfra","dossier_json_url":"https://onlylabs.fyi/labs/deepinfra/dossier.json"},"related_urls":{"signal":"https://onlylabs.fyi/signals/2a903af5-9d24-4c22-b727-ad90821882dd","signal_json":"https://onlylabs.fyi/signals/2a903af5-9d24-4c22-b727-ad90821882dd/signal.json","source":"https://github.com/deepinfra/flash-attention","lab_dossier":"https://onlylabs.fyi/labs/deepinfra","lab_dossier_json":"https://onlylabs.fyi/labs/deepinfra/dossier.json","analysis":"https://onlylabs.fyi/analysis/deepinfra","analysis_json":"https://onlylabs.fyi/analysis/deepinfra/analysis.json","analysis_evidence_json":"https://onlylabs.fyi/analysis/deepinfra/evidence.json","category":"https://onlylabs.fyi/neoclouds","category_json":"https://onlylabs.fyi/neoclouds.json","category_feed":"https://onlylabs.fyi/neoclouds/feed.xml","category_signals_json":"https://onlylabs.fyi/signals.json?category=neocloud","topic":"https://onlylabs.fyi/topics/forks","topic_signals_json":"https://onlylabs.fyi/topics/forks/signals.json?category=neocloud","topic_feed":"https://onlylabs.fyi/topics/forks/feed.xml?category=neocloud","data_business":null},"answer_pack":{"answer":"DeepInfra forked deepinfra/flash-attention (forked from vllm-project/flash-attention). This fork signal points to upstream code the lab may be inspecting, patching, or building on. High-signal details: repo deepinfra/flash-attention · parent vllm-project/flash-attention · Routine fork of a notable repo. onlylabs links this event to 1 captured evidence page and 6 related fork signals.","signal_desk":"forks","source_context":{"source_url":"https://github.com/deepinfra/flash-attention","source_host":"github.com","occurred_at":"2025-02-19T22:18:06+00:00","first_seen_at":"2026-06-05T22:32:12.181599+00:00","date_source":"source","context":"forked from vllm-project/flash-attention"},"context_markers":[{"label":"Lab","value":"DeepInfra","source":"signal"},{"label":"Signal desk","value":"forks","source":"signal"},{"label":"Source host","value":"github.com","source":"source"},{"label":"Repository","value":"deepinfra/flash-attention","source":"source"},{"label":"Parent","value":"vllm-project/flash-attention","source":"source"},{"label":"Notability","value":"Routine fork of a notable repo","source":"signal"},{"label":"Watch term","value":"Eval methodology","source":"evidence"},{"label":"Watch term","value":"Infrastructure","source":"evidence"}],"evidence_coverage":{"target_pages":1,"captured_pages":1,"readable_pages":1,"capture_methods":["plain"],"missing_page_urls":[],"failed_page_urls":[],"blocked_page_urls":[],"page_urls":["https://github.com/deepinfra/flash-attention"],"related_signals":6,"has_source_url":true,"latest_page_fetched_at":"2026-06-11T02:54:23.295206+00:00"},"data_business":{"matches":false,"lanes":[],"matched_terms":[],"score":null,"reason":null},"agent_handoff":{"signal_json":"https://onlylabs.fyi/signals/2a903af5-9d24-4c22-b727-ad90821882dd/signal.json","dossier_json":"https://onlylabs.fyi/labs/deepinfra/dossier.json","analysis_json":"https://onlylabs.fyi/analysis/deepinfra/analysis.json","analysis_evidence_json":"https://onlylabs.fyi/analysis/deepinfra/evidence.json","topic_signals_json":"https://onlylabs.fyi/topics/forks/signals.json?category=neocloud","topic_feed":"https://onlylabs.fyi/topics/forks/feed.xml?category=neocloud","category_signals_json":"https://onlylabs.fyi/signals.json?category=neocloud","data_radar_json":null,"opportunities_json":null},"analysis_playbook":{"objective":"Turn fork activity into a map of upstream tools, libraries, model utilities, agent frameworks, and research dependencies being inspected or adapted.","evidence_focus":["forked parent repo","repo owner","language","description","stars","org and recency"],"extraction_questions":["Which upstream projects are repeated across labs?","Which forks point to evals, agents, infrastructure, model tooling, or data pipelines?","Which low-profile forks could precede a release or hiring wave?","Which labs are converging on the same technical adjacency?"],"signal_questions":["What upstream dependency or technical adjacency does this fork reveal?","Which upstream projects are repeated across labs?","Which forks point to evals, agents, infrastructure, model tooling, or data pipelines?","Do the 6 related fork signals show a repeated pattern?"],"output_fields":["org","upstream_repo","language","technical_theme","evidence_url"],"data_business_relevance":"Data-business lane extraction is scoped to frontier labs; for this category, keep conclusions tied to category-specific strategy, source evidence, and follow-up questions.","required_sources":[{"label":"signal_json","url":"https://onlylabs.fyi/signals/2a903af5-9d24-4c22-b727-ad90821882dd/signal.json","required":true},{"label":"source","url":"https://github.com/deepinfra/flash-attention","required":true},{"label":"dossier_json","url":"https://onlylabs.fyi/labs/deepinfra/dossier.json","required":true},{"label":"analysis_evidence_json","url":"https://onlylabs.fyi/analysis/deepinfra/evidence.json","required":true},{"label":"topic_signals_json","url":"https://onlylabs.fyi/topics/forks/signals.json?category=neocloud","required":false},{"label":"data_radar_json","url":null,"required":false}],"expected_output":["one-paragraph source-grounded interpretation","category-specific implication","confidence and missing evidence","recommended next source to inspect"],"prompt_seed":"Using only the linked onlylabs JSON, captured source context, and cited evidence, analyze DeepInfra's fork signal \"deepinfra/flash-attention\" for neocloud strategy."},"semantic_triples":[{"subject":"DeepInfra","predicate":"forked","object":"deepinfra/flash-attention","text":"DeepInfra forked deepinfra/flash-attention."},{"subject":"deepinfra/flash-attention","predicate":"is classified as","object":"fork signal","text":"deepinfra/flash-attention is classified as fork signal."},{"subject":"deepinfra/flash-attention","predicate":"belongs to","object":"forks desk","text":"deepinfra/flash-attention belongs to forks desk."},{"subject":"deepinfra/flash-attention","predicate":"has context","object":"forked from vllm-project/flash-attention","text":"deepinfra/flash-attention has context forked from vllm-project/flash-attention."},{"subject":"deepinfra/flash-attention","predicate":"has evidence coverage","object":"1 captured evidence page","text":"deepinfra/flash-attention has evidence coverage 1 captured evidence page."},{"subject":"deepinfra/flash-attention","predicate":"has captured page count","object":"1","text":"deepinfra/flash-attention has captured page count 1."},{"subject":"deepinfra/flash-attention","predicate":"has readable page count","object":"1","text":"deepinfra/flash-attention has readable page count 1."},{"subject":"deepinfra/flash-attention","predicate":"has related signal count","object":"6","text":"deepinfra/flash-attention has related signal count 6."},{"subject":"deepinfra/flash-attention","predicate":"has analysis playbook objective","object":"Turn fork activity into a map of upstream tools, libraries, model utilities, agent frameworks, and research dependencies being inspected or adapted.","text":"deepinfra/flash-attention has analysis playbook objective Turn fork activity into a map of upstream tools, libraries, model utilities, agent frameworks, and research dependencies being inspected or adapted.."},{"subject":"deepinfra/flash-attention","predicate":"has source host","object":"github.com","text":"deepinfra/flash-attention has source host github.com."},{"subject":"deepinfra/flash-attention","predicate":"has lab","object":"DeepInfra","text":"deepinfra/flash-attention has lab DeepInfra."},{"subject":"deepinfra/flash-attention","predicate":"has signal desk","object":"forks","text":"deepinfra/flash-attention has signal desk forks."},{"subject":"deepinfra/flash-attention","predicate":"has source host","object":"github.com","text":"deepinfra/flash-attention has source host github.com."},{"subject":"deepinfra/flash-attention","predicate":"has repository","object":"deepinfra/flash-attention","text":"deepinfra/flash-attention has repository deepinfra/flash-attention."},{"subject":"deepinfra/flash-attention","predicate":"has parent","object":"vllm-project/flash-attention","text":"deepinfra/flash-attention has parent vllm-project/flash-attention."},{"subject":"deepinfra/flash-attention","predicate":"has notability","object":"Routine fork of a notable repo","text":"deepinfra/flash-attention has notability Routine fork of a notable repo."},{"subject":"deepinfra/flash-attention","predicate":"has watch term","object":"Eval methodology","text":"deepinfra/flash-attention has watch term Eval methodology."},{"subject":"deepinfra/flash-attention","predicate":"has watch term","object":"Infrastructure","text":"deepinfra/flash-attention has watch term Infrastructure."}]},"intelligence":{"signal_desk":"forks","answer":"DeepInfra forked deepinfra/flash-attention (forked from vllm-project/flash-attention). This fork signal points to upstream code the lab may be inspecting, patching, or building on. High-signal details: repo deepinfra/flash-attention · parent vllm-project/flash-attention · Routine fork of a notable repo. onlylabs links this event to 1 captured evidence page and 6 related fork signals.","semantic_triples":[{"subject":"DeepInfra","predicate":"forked","object":"deepinfra/flash-attention","text":"DeepInfra forked deepinfra/flash-attention."},{"subject":"deepinfra/flash-attention","predicate":"is classified as","object":"fork signal","text":"deepinfra/flash-attention is classified as fork signal."},{"subject":"deepinfra/flash-attention","predicate":"belongs to","object":"forks desk","text":"deepinfra/flash-attention belongs to forks desk."},{"subject":"deepinfra/flash-attention","predicate":"has context","object":"forked from vllm-project/flash-attention","text":"deepinfra/flash-attention has context forked from vllm-project/flash-attention."},{"subject":"deepinfra/flash-attention","predicate":"has evidence coverage","object":"1 captured evidence page","text":"deepinfra/flash-attention has evidence coverage 1 captured evidence page."}]},"signal":{"id":"2a903af5-9d24-4c22-b727-ad90821882dd","url":"https://onlylabs.fyi/signals/2a903af5-9d24-4c22-b727-ad90821882dd","json_url":"https://onlylabs.fyi/signals/2a903af5-9d24-4c22-b727-ad90821882dd/signal.json","source_url":"https://github.com/deepinfra/flash-attention","title":"deepinfra/flash-attention","summary":"DeepInfra forked upstream code. onlylabs watches forks for tooling, infrastructure, research dependencies, and product-adjacent work.","context":"forked from vllm-project/flash-attention","kind":{"key":"repo_forked","label":"Fork"},"org":{"slug":"deepinfra","name":"DeepInfra","category":"neocloud"},"occurred_at":"2025-02-19T22:18:06+00:00","first_seen_at":"2026-06-05T22:32:12.181599+00:00","date_source":"source","evidence_coverage":{"target_pages":1,"captured_pages":1,"readable_pages":1,"capture_methods":["plain"],"missing_page_urls":[],"failed_page_urls":[],"blocked_page_urls":[],"page_urls":["https://github.com/deepinfra/flash-attention"]},"facets":{"repo":"deepinfra/flash-attention","parent":"vllm-project/flash-attention"},"traction":{"github_stars":0,"hn_points":null,"hn_comments":null,"hn_story_id":null,"hf_downloads":null,"hf_likes":null},"data_radar":null},"primary_evidence_page":{"url":"https://github.com/deepinfra/flash-attention","final_url":"https://github.com/deepinfra/flash-attention","title":"deepinfra/flash-attention repository metadata","http_status":200,"content_type":"application/json","capture_method":"plain","fetched_at":"2026-06-11T02:54:23.295206+00:00","bytes":42455,"raw_path":"134beb4312898db7a43474a1744d6fc9a203b64fffed749c3a4a2f070fcc59b9.json","content_hash":"2cf2522622d3d7a331f40c2945d65afd63158ca405c780dc9a2d60873258f171","excerpt_chars":1200,"truncated":true,"excerpt":"deepinfra/flash-attention Description: Fast and memory-efficient exact attention Language: Python License: BSD-3-Clause Stars: 0 Forks: 0 Open issues: 0 Created: 2025-02-19T22:18:06Z Pushed: 2025-02-20T00:29:54Z Default branch: main Fork: yes Parent repository: vllm-project/flash-attention Archived: no README: FlashAttention This repository provides the official implementation of FlashAttention and FlashAttention-2 from the following papers. **FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness** Tri Dao, Daniel Y. Fu, Stefano Ermon, Atri Rudra, Christopher Ré Paper: https://arxiv.org/abs/2205.14135 IEEE Spectrum [article](https://spectrum.ieee.org/mlperf-rankings-2022) about our submission to the MLPerf 2.0 benchmark using FlashAttention. ![FlashAttention](assets/flashattn_banner.jpg) **FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning** Tri Dao Paper: https://tridao.me/publications/flash2/flash2.pdf ![FlashAttention-2](assets/flashattention_logo.png) Usage We've been very happy to see FlashAttention being widely adopted in such a short time after its release. This..."},"evidence_pages":[{"url":"https://github.com/deepinfra/flash-attention","final_url":"https://github.com/deepinfra/flash-attention","title":"deepinfra/flash-attention repository metadata","http_status":200,"content_type":"application/json","capture_method":"plain","fetched_at":"2026-06-11T02:54:23.295206+00:00","bytes":42455,"raw_path":"134beb4312898db7a43474a1744d6fc9a203b64fffed749c3a4a2f070fcc59b9.json","content_hash":"2cf2522622d3d7a331f40c2945d65afd63158ca405c780dc9a2d60873258f171","excerpt_chars":1200,"truncated":true,"excerpt":"deepinfra/flash-attention Description: Fast and memory-efficient exact attention Language: Python License: BSD-3-Clause Stars: 0 Forks: 0 Open issues: 0 Created: 2025-02-19T22:18:06Z Pushed: 2025-02-20T00:29:54Z Default branch: main Fork: yes Parent repository: vllm-project/flash-attention Archived: no README: FlashAttention This repository provides the official implementation of FlashAttention and FlashAttention-2 from the following papers. **FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness** Tri Dao, Daniel Y. Fu, Stefano Ermon, Atri Rudra, Christopher Ré Paper: https://arxiv.org/abs/2205.14135 IEEE Spectrum [article](https://spectrum.ieee.org/mlperf-rankings-2022) about our submission to the MLPerf 2.0 benchmark using FlashAttention. ![FlashAttention](assets/flashattn_banner.jpg) **FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning** Tri Dao Paper: https://tridao.me/publications/flash2/flash2.pdf ![FlashAttention-2](assets/flashattention_logo.png) Usage We've been very happy to see FlashAttention being widely adopted in such a short time after its release. This..."}],"related_signals":[{"id":"6ff8fada-5c71-4734-89ac-0152c9382688","url":"https://onlylabs.fyi/signals/6ff8fada-5c71-4734-89ac-0152c9382688","source_url":"https://github.com/deepinfra/TorchSpec","title":"deepinfra/TorchSpec","context":"forked from lightseekorg/TorchSpec","kind":{"key":"repo_forked","label":"Fork"},"org":{"slug":"deepinfra","name":"DeepInfra","category":"neocloud"},"occurred_at":"2026-06-04T22:52:45+00:00","first_seen_at":"2026-06-05T22:32:12.181599+00:00","date_source":"source"},{"id":"d0c20615-2b18-4fd2-aa9a-391757002eb0","url":"https://onlylabs.fyi/signals/d0c20615-2b18-4fd2-aa9a-391757002eb0","source_url":"https://github.com/deepinfra/tiktoken","title":"deepinfra/tiktoken","context":"forked from openai/tiktoken","kind":{"key":"repo_forked","label":"Fork"},"org":{"slug":"deepinfra","name":"DeepInfra","category":"neocloud"},"occurred_at":"2026-03-06T23:43:26+00:00","first_seen_at":"2026-06-05T22:32:12.181599+00:00","date_source":"source"},{"id":"6cf91d48-6d66-461b-bac2-07963d810083","url":"https://onlylabs.fyi/signals/6cf91d48-6d66-461b-bac2-07963d810083","source_url":"https://github.com/deepinfra/vllm-omni","title":"deepinfra/vllm-omni","context":"forked from vllm-project/vllm-omni","kind":{"key":"repo_forked","label":"Fork"},"org":{"slug":"deepinfra","name":"DeepInfra","category":"neocloud"},"occurred_at":"2026-03-06T21:15:02+00:00","first_seen_at":"2026-06-05T22:32:12.181599+00:00","date_source":"source"},{"id":"ef4a4eca-98dc-4a61-b264-2861da8679c0","url":"https://onlylabs.fyi/signals/ef4a4eca-98dc-4a61-b264-2861da8679c0","source_url":"https://github.com/deepinfra/hub-docs","title":"deepinfra/hub-docs","context":"forked from huggingface/hub-docs","kind":{"key":"repo_forked","label":"Fork"},"org":{"slug":"deepinfra","name":"DeepInfra","category":"neocloud"},"occurred_at":"2026-02-24T18:59:21+00:00","first_seen_at":"2026-06-05T22:32:12.181599+00:00","date_source":"source"},{"id":"8b01be6b-1201-4717-a6e4-27fa1786296a","url":"https://onlylabs.fyi/signals/8b01be6b-1201-4717-a6e4-27fa1786296a","source_url":"https://github.com/deepinfra/dynamo","title":"deepinfra/dynamo","context":"forked from ai-dynamo/dynamo","kind":{"key":"repo_forked","label":"Fork"},"org":{"slug":"deepinfra","name":"DeepInfra","category":"neocloud"},"occurred_at":"2026-01-26T23:19:52+00:00","first_seen_at":"2026-06-05T22:32:12.181599+00:00","date_source":"source"},{"id":"aaeffd61-b0e7-46d8-9736-093ec46d0805","url":"https://onlylabs.fyi/signals/aaeffd61-b0e7-46d8-9736-093ec46d0805","source_url":"https://github.com/deepinfra/Model-Optimizer","title":"deepinfra/Model-Optimizer","context":"forked from NVIDIA/Model-Optimizer","kind":{"key":"repo_forked","label":"Fork"},"org":{"slug":"deepinfra","name":"DeepInfra","category":"neocloud"},"occurred_at":"2026-01-12T18:01:11+00:00","first_seen_at":"2026-06-05T22:32:12.181599+00:00","date_source":"source"}]}