Lightning-AI/litgpt v0.5.12

v0.5.12

Repository: Lightning-AI/litgpt

Tag: v0.5.12

Published: 2025-12-18T15:51:48Z

Prerelease: no

Release notes:

What's Changed

• fix: update TinyStories optimize to use item_loader=TokensLoader() by @bhimrazy in https://github.com/Lightning-AI/litgpt/pull/2148
• [FEAT] Add Grouped Topk Routing to LLaMAMoE (Based on DeepseekV3MoE) by @ysjprojects in https://github.com/Lightning-AI/litgpt/pull/2134
• Add timeout parameter to run_server. by @hv10 in https://github.com/Lightning-AI/litgpt/pull/2157
• fix: chat prompt lowercasing bug by @bhimrazy in https://github.com/Lightning-AI/litgpt/pull/2155
• chore: Drop Python 3.9 support from ci by @bhimrazy in https://github.com/Lightning-AI/litgpt/pull/2161
• fix: handle gemma 3 weights prefix during hf conversion by @scascar in https://github.com/Lightning-AI/litgpt/pull/2156
• Fix to avoid RuntimeError by @mseeger in https://github.com/Lightning-AI/litgpt/pull/2138
• fix: pin lm-eval<0.4.9.1 for trust_remote_code issue by @bhimrazy in https://github.com/Lightning-AI/litgpt/pull/2168
• Small refactorings by @mseeger in https://github.com/Lightning-AI/litgpt/pull/2166
• docs: enhance Python API tutorial with fixes and new section by @carsch18 in https://github.com/Lightning-AI/litgpt/pull/2169
• fix: sliding window KV cache for Gemma-3 models (issue #2145) by @Chelsi-create in https://github.com/Lightning-AI/litgpt/pull/2153

New Contributors

• @lianakoleva made their first contribution in https://github.com/Lightning-AI/litgpt/pull/2132
• @hv10 made their first contribution in https://github.com/Lightning-AI/litgpt/pull/2157
• @scascar made their first contribution in https://github.com/Lightning-AI/litgpt/pull/2156
• @carsch18 made their first contribution in https://github.com/Lightning-AI/litgpt/pull/2169
• @Chelsi-create made their first contribution in https://github.com/Lightning-AI/litgpt/pull/2153

Full Changelog: https://github.com/Lightning-AI/litgpt/compare/v0.5.11...v0.5.12