Qiaolin Yu d6898dd253 Add return hidden state in the native API (#3897) ...

Co-authored-by: Beichen-Ma <mabeichen12@gmail.com>
Co-authored-by: Chayenne <zhaochen20@outlook.com>

2025-02-26 22:06:54 -08:00

..

configs

Refactor flashinfer logic for deepseek v3 and fix accuracy bug (#3785)

2025-02-24 04:07:25 -08:00

constrained

[Feature] Support llguidance for constrained decoding (#3298)

2025-02-26 10:41:49 -08:00

distributed

Fix allgather ops inside cuda graphs (#3709)

2025-02-25 08:39:10 -08:00

entrypoints

Improve: fix typos (#3801)

2025-02-24 16:51:23 -08:00

layers

[Rocm] Fix to the rocm_mla_decode_rope.py returning random result (#3898)

2025-02-26 17:05:30 -08:00

lora

[Fix] Fix bugs and refactor codes in lora for better scalability. (#3652)

2025-02-20 11:51:57 -08:00

managers

Add return hidden state in the native API (#3897)

2025-02-26 22:06:54 -08:00

mem_cache

Hierarchical Caching for SGLang (#2693)

2025-02-23 21:56:30 -08:00

metrics

Improve metrics, logging, and importing orders (#2992)

2025-01-19 18:36:59 -08:00

model_executor

Add return hidden state in the native API (#3897)

2025-02-26 22:06:54 -08:00

model_loader

Add support for nvidia modelopt fp8 kv cache (#3223)

2025-02-22 07:04:58 +08:00

models

Fix allgather ops inside cuda graphs (#3709)

2025-02-25 08:39:10 -08:00

openai_api

Revert "Rename TokenizerManager to StdOrchestrator" (#3828)

2025-02-24 14:47:59 -08:00

sampling

Add return hidden state in the native API (#3897)

2025-02-26 22:06:54 -08:00

speculative

Support NextN (MTP) speculative decoding for DeepSeek-V3/R1 (#3582)

2025-02-15 05:28:34 +08:00

_custom_ops.py

Roll back to use vllm custom allreduce (#3006)

2025-01-20 04:03:15 -08:00

aio_rwlock.py

A better aio rwlock that guarantees the order (#2547)

2024-12-22 15:44:32 -08:00

conversation.py

[Feature] Support minicpmv v2.6 (#2785)

2025-01-18 14:14:19 -08:00

custom_op.py

Use forward_cuda to execute custom op for hip platform (#3305)

2025-02-05 02:58:17 +00:00

function_call_parser.py

Fix: Runtime error for function calling (#3300)

2025-02-06 20:52:01 -08:00

hf_transformers_utils.py

Model: Support Qwen 2.5 vl (#3258)

2025-02-16 00:58:53 -08:00

mm_utils.py

docs: fix module docstrings and copyright headers (#2077)

2024-11-22 22:16:53 +08:00

model_parallel.py

Make torch TP composable with torchao (#2436)

2024-12-11 04:21:42 -08:00

server_args.py

Add return hidden state in the native API (#3897)

2025-02-26 22:06:54 -08:00

server.py

Misc fixes for eagle (flush_cache, CPU overhead) (#3014)

2025-01-20 20:27:38 -08:00

torch_memory_saver_adapter.py

Add a new api configure_logging to allow dumping the requests (#2875)

2025-01-13 14:24:00 -08:00

utils.py

[Bug]: Fix maximum recursion depth triggered on exception exit (#3519)

2025-02-25 09:39:38 -08:00