GitHub / FasterDecoding/Medusa / commits
Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads
| SHA | Message | Author | Date | Stats |
|---|---|---|---|---|
| e2a5d20c | merge Merge pull request #97 from Narsil/medusa2 |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
almost 2 years ago | |
| 2b8020f2 | Creating medusa2. | Nicolas Patry <p****s@p****m> | almost 2 years ago | |
| 5e980538 | merge Merge pull request #83 from Narsil/recipe_for_other_models |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
about 2 years ago | |
| 64f49242 | Update create_data.py |
Nicolas Patry <p****s@p****m>
Committed by: GitHub <n****y@g****m> |
about 2 years ago | |
| 6e713f46 | Forgot the deepspeed config. | Nicolas Patry <p****s@p****m> | about 2 years ago | |
| 0bfdcd23 | Adding recipe for other models (non llama, non vicuna). | Nicolas Patry <p****s@p****m> | about 2 years ago | |
| 700ff848 | update README and add back legacy code for compatibility | leeyeehoo <t****i@p****u> | about 2 years ago | |
| 93bee11f | merge Merge pull request #72 from zhyncs/patch-1 |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
about 2 years ago | |
| d3464f61 | update Community Adoption for RTP-LLM |
zhyncs <m****e@z****m>
Committed by: GitHub <n****y@g****m> |
about 2 years ago | |
| fc0a6c7a | merge Merge pull request #71 from FasterDecoding/v1.0-prerelease |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
about 2 years ago | |
| bd95225a | fix | leeyeehoo <t****i@p****u> | about 2 years ago | |
| 4c291e45 | release | leeyeehoo <t****i@p****u> | about 2 years ago | |
| 95f1271b | medusa choice auto dispatch | leeyeehoo <t****i@p****u> | about 2 years ago | |
| a4ec58e2 | fix eval | leeyeehoo <l****0@o****m> | about 2 years ago | |
| 1068af11 | update eval | leeyeehoo <l****0@o****m> | about 2 years ago | |
| f22d72fa | solve the compatibility w v0.1 | leeyeehoo <l****0@o****m> | about 2 years ago | |
| e1154c00 | update choices | leeyeehoo <l****0@o****m> | about 2 years ago | |
| 2b898144 | update cli | leeyeehoo <l****0@o****m> | about 2 years ago | |
| ac75468a | data generation for self-distillation | Tianle Cai <l****0@o****m> | about 2 years ago | |
| 0ac14da9 | add extra sampling strategies | Tianle Cai <l****0@o****m> | about 2 years ago | |
| 1facb55d | support mistral | Tianle Cai <l****0@o****m> | about 2 years ago | |
| a3578952 | merge Merge branch 'v1.0-prerelease' of github.com:FasterDecoding/Medusa into v1.0-... | leeyeehoo <l****0@o****m> | over 2 years ago | |
| af8c7d95 | add mistral | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 46f911eb | update model loading APIs | Tianle Cai <t****e@c****i> | over 2 years ago | |
| 6294228e | update medusa eval | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 15934823 | update gitignore | leeyeehoo <l****0@o****m> | over 2 years ago | |
| dd9c8a53 | merge Merge pull request #53 from FasterDecoding/llm_judge |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 55742873 | baseline | leeyeehoo <l****0@o****m> | over 2 years ago | |
| ccb95039 | baseline | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 0061b134 | fix bug | HarveyP123 <h****7@g****m> | over 2 years ago | |
| 077977a5 | merge Merge pull request #46 from FasterDecoding/readme_update |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| c1783036 | update roadmap | HarveyP123 <h****7@g****m> | over 2 years ago | |
| 60eeab92 | fix a cli bug | HarveyP123 <h****7@g****m> | over 2 years ago | |
| 10a2697c | merge Merge pull request #42 from FasterDecoding/sparse_tree |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 89f8ec04 | merge Merge branch 'main' into sparse_tree |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 5d374e9b | upload train accuracy | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 8888d3a3 | add judge results | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 4681013c | add judge results | leeyeehoo <l****0@o****m> | over 2 years ago | |
| e315780a | update notebook | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 8ce8fa9f | merge Merge pull request #38 from FasterDecoding/ctlllll-patch-1 |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 03f3c0d4 | add development bounty |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| d690f7fa | merge Merge pull request #26 from Btlmd/main |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 6264fe85 | merge Merge pull request #27 from rajveer43/patch-1 |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 8891f351 | merge Merge pull request #23 from Mrw33554432/main |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 68b50404 | fix training base model config | lambda <j****7@h****m> | over 2 years ago | |
| 5b3d2b88 | add docstgins | Rockerz <6****3@u****m> | over 2 years ago | |
| e5cdff32 | add base model override | lambda <j****7@h****m> | over 2 years ago | |
| 0c55d526 | add a model loader page and some settings. | Mrw33554432 <s****9@o****m> | over 2 years ago | |
| af86f3b2 | update notebook | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 55c77cba | update notebook | leeyeehoo <l****0@o****m> | over 2 years ago | |
| e9d21919 | merge Merge branch 'sparse_tree' of github.com:FasterDecoding/Medusa into sparse_tree | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 4054fa25 | Revert "Add sparse tree" | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 922689a1 | Add sparse tree | leeyeehoo <l****0@o****m> | over 2 years ago | |
| 11af0aa5 | Add sparse tree | HarveyP123 <h****7@g****m> | over 2 years ago | |
| b27b6fd6 | Add a simple gradio interface, make life easier | Mrw33554432 <s****9@o****m> | over 2 years ago | |
| 269d2a5b | merge Merge pull request #22 from FasterDecoding/main |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| c4e0c63d | merge Merge pull request #21 from FasterDecoding/leeyeehoo-patch-1 |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| cc5192f2 | Update ROADMAP.md |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 1e07b6ca | merge Merge pull request #20 from FasterDecoding/sparse_tree |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| ae679911 | merge Merge pull request #19 from FasterDecoding/main |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 0d4a166f | merge Merge pull request #17 from ctlllll/main |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 70577b74 | Fix save medusa_head | Tianle Cai <t****i@p****u> | over 2 years ago | |
| b50cb412 | merge Merge pull request #15 from FasterDecoding/git-lfs |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| bf0614d8 | Add git-lfs instruction |
Tianle Cai <t****i@p****u>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| e76b4800 | merge Merge pull request #10 from eltociear/patch-1 |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 6d0f34ab | Update ROADMAP.md |
Ikko Eltociear Ashimine <e****r@g****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 90029aa4 | merge Merge pull request #4 from kalomaze/patch-1 |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 6815d3f2 | Cleaned up README |
kalomaze <6****e@u****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 234aa339 | Update README.md |
Yuhong Li <l****0@o****m>
Committed by: GitHub <n****y@g****m> |
over 2 years ago | |
| 10e73877 | modify readme description | Tianle Cai <t****i@p****u> | over 2 years ago | |
| 1438d1ba | fix cff file | Tianle Cai <t****i@p****u> | over 2 years ago | |
| ab5d60c8 | initial release | Tianle Cai <t****i@p****u> | over 2 years ago |