Lianmin Zheng 9e1014cf99 Revert "Add fast decode plan for flashinfer mla" (#4008)

2025-03-02 19:29:10 -08:00

..

configs

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

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

constrained

[Feature] integrate Structural Tag in xgrammar backend for function calling (#3566)

2025-02-27 23:33:41 -08:00

distributed

Enable custom AR for AMD GPUs and maintain it in sgl-kernel (#3406)

2025-03-02 15:19:06 -08:00

entrypoints

Refactor: Move return_hidden_states to the generate input (#3985)

2025-03-01 17:51:29 -08:00

layers

Revert "Add fast decode plan for flashinfer mla" (#4008)

2025-03-02 19:29:10 -08:00

lora

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

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

managers

Revert "Add fast decode plan for flashinfer mla" (#4008)

2025-03-02 19:29:10 -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

Revert "Add fast decode plan for flashinfer mla" (#4008)

2025-03-02 19:29:10 -08:00

model_loader

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

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

models

[feat] add small vocab table for eagle's draft model[1]. (#3822)

2025-03-02 18:58:45 -08:00

openai_api

[Feature] integrate Structural Tag in xgrammar backend for function calling (#3566)

2025-02-27 23:33:41 -08:00

sampling

Refactor: Move return_hidden_states to the generate input (#3985)

2025-03-01 17:51:29 -08:00

speculative

[feat] add small vocab table for eagle's draft model[1]. (#3822)

2025-03-02 18:58:45 -08:00

_custom_ops.py

Enable custom AR for AMD GPUs and maintain it in sgl-kernel (#3406)

2025-03-02 15:19:06 -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

[feat] add small vocab table for eagle's draft model[1]. (#3822)

2025-03-02 18:58:45 -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

[Feature] SPMD for SGLang + Verl (#3852)

2025-02-28 09:53:10 -08:00