Mingliang Li d007387aa7 [Bugfix] Cache added_vocab to avoid per-token overhead (#30351) ...

Signed-off-by: limingliang <limingliang@stepfun.com>
Co-authored-by: limingliang <limingliang@stepfun.com>

2025-12-10 12:05:51 +08:00

..

__init__.py

[Frontend] supports deepseekv32 chat template (#29837)

2025-12-03 20:53:44 +08:00

deepseek_v32_encoding.py

[Bugfix] fixed deepseekv32 tool calling error (#30025)

2025-12-04 15:12:12 +08:00

deepseekv32.py

[Bugfix] Cache added_vocab to avoid per-token overhead (#30351)

2025-12-10 12:05:51 +08:00

detokenizer_utils.py

[Chore] Move detokenizer_utils to vllm/tokenizers (#29727)

2025-11-29 06:25:17 -08:00

hf.py

[Misc] Unify tokenizer registration (#29767)

2025-12-01 11:34:58 +00:00

mistral.py

[Bugfix]: Fix TokenizerLike interface (#30009)

2025-12-05 20:56:40 -08:00

protocol.py

[Bugfix]: Fix TokenizerLike interface (#30009)

2025-12-05 20:56:40 -08:00

registry.py

Revert "[Renderer] Separate out RendererConfig from ModelConfig (#30145)" (#30199)

2025-12-07 00:00:22 -08:00