Yineng Zhang b1a540ec42 feat: update GemmaRMSNorm (#1232)

2024-08-28 22:47:34 +10:00

..

constrained

[FEAT] JSON constrained support (#1125)

2024-08-26 09:37:26 -07:00

layers

feat: update GemmaRMSNorm (#1232)

2024-08-28 22:47:34 +10:00

managers

hotfix: revert sampler CUDA Graph (#1242)

2024-08-28 21:16:47 +10:00

mem_cache

[Feature] Support fp8 e5m2 kv cache with flashinfer (#1204)

2024-08-25 17:38:11 -07:00

model_executor

hotfix: revert sampler CUDA Graph (#1242)

2024-08-28 21:16:47 +10:00

models

feat: update GemmaRMSNorm (#1232)

2024-08-28 22:47:34 +10:00

openai_api

[FEAT] Support batches cancel (#1222)

2024-08-26 23:28:26 +00:00

sampling

hotfix: revert sampler CUDA Graph (#1242)

2024-08-28 21:16:47 +10:00

conversation.py

[Feat/WIP] add llava-onevision, with support for (1) siglip encoder, (2) qwen2 decoder (3) openai api compatible server. (#1123)

2024-08-23 14:11:16 -07:00

hf_transformers_utils.py

[Minor] Improve the function organization in TokenizerManager & improve loggers (#1208)

2024-08-25 14:46:34 -07:00

mm_utils.py

[Feat/WIP] add llava-onevision, with support for (1) siglip encoder, (2) qwen2 decoder (3) openai api compatible server. (#1123)

2024-08-23 14:11:16 -07:00

model_config.py

Support MLA for DeepSeek-V2 with Triton - step 1 (#905)

2024-08-05 03:40:33 +10:00

server_args.py

[Feature] Support fp8 e5m2 kv cache with flashinfer (#1204)

2024-08-25 17:38:11 -07:00

server.py

feat: support sm75 with FlashInfer v0.1.6 (#1233)

2024-08-28 18:39:12 +10:00

utils.py

[Minor] Improve the function organization in TokenizerManager & improve loggers (#1208)

2024-08-25 14:46:34 -07:00