sglang

batch_invariant_ops

Support fast gemm when in batch invariant DeepGEMM fallback (#13259 )

2025-11-15 16:34:15 +08:00

batch_overlap

[Feature] Xiaomi MiMo-V2-Flash day0 support (#15207 )

2025-12-19 11:40:07 +08:00

checkpoint_engine

[Ckpt Engine] feat: new sglang entrypoint support for update (#12216 )

2025-10-30 10:39:27 +08:00

compilation

Custom All Reduce for Piecewise Cuda Graph (#15356 )

2025-12-25 23:55:15 +08:00

configs

Separate swa and local attention chunk cache eviction (#15820 )

2025-12-26 09:34:22 +08:00

connector

[Lint] Add python/sglang to ruff F401 checks and remove unused imports in files (#11685 )

2025-10-17 16:49:46 -07:00

constrained

Adding tool calling and reasoning parser support for Intern-S1 (#14866 )

2025-12-16 00:01:58 -08:00

debug_utils

Super tiny add more info in dumper (#14145 )

2025-11-30 09:07:39 +08:00

disaggregation

Fix temp_prefill_info assertion error in PP disaggregation mode (#15943 )

2025-12-27 22:46:55 +08:00

distributed

[VLM] Support ViT Piecewise CUDA Graph for Qwen3-VL (#15320 )

2025-12-20 21:00:07 +08:00

dllm

Tiny extract ModelRunnerOutput (#15400 )

2025-12-18 22:18:45 +08:00

elastic_ep

[2/N] Added the core structure of elastic EP and the eplb algorithm with faulty rank (#10606 )

2025-10-22 01:13:31 -07:00

entrypoints

[model-gateway]: fix grpc embedding test (#15934 )

2025-12-27 05:56:54 -05:00

eplb

Support EPLB balancedness prometheus metric without GPU->CPU synchronize (#15401 )

2025-12-18 22:24:23 +08:00

function_call

[Tool Call][DSV32] Streamline function call parameters (#14750 )

2025-12-26 11:35:30 -08:00

grpc

[model-gateway]: add gRPC router embeddings endpoint implementation (#15273 )

2025-12-23 09:12:22 -08:00

hardware_backend/npu

[NPU] Support w4a8 with activation clip (#14736 )

2025-12-27 16:19:46 +08:00

layers

[NPU] Support w4a8 with activation clip (#14736 )

2025-12-27 16:19:46 +08:00

lora

Feature/Fix multi lora scheduler blocking issue and evict LoRA None lastly (#14795 )

2025-12-12 17:13:05 -08:00

managers

Add a test case for crash dump (#15905 )

2025-12-26 19:39:28 -08:00

mem_cache

[BUGFIX] fix edge case for qwen3-next (#14209 )

2025-12-26 17:14:37 +08:00

metrics

Add LoRA metrics for potential auto scaling (#15149 )

2025-12-24 12:13:52 -08:00

model_executor

Introduce ModelRunnerKVCacheMixin to simplify the code. (#15821 )

2025-12-26 11:17:33 +08:00

model_loader

[Feature] support fastsafetensors (#15091 )

2025-12-23 22:33:56 +08:00

models

[BugFix][VLM] Correct weight loading with tie_word_embeddings == False (#15398 )

2025-12-26 17:30:50 -08:00

multimodal

Revert "[VLM] Refactor load_mm_data to improve performance" (#15911 )

2025-12-26 13:43:18 -08:00

multiplex

hotfix: add type hints to scheduler mixins (#15916 )

2025-12-26 17:08:11 -08:00

parser

fix: prevent points regex from matching checkpoints/endpoints (#15120 )

2025-12-17 12:38:18 -08:00

sampling

Add customized sampler registration (#15423 )

2025-12-18 23:10:23 -08:00

speculative

Separate swa and local attention chunk cache eviction (#15820 )

2025-12-26 09:34:22 +08:00

tokenizer

[Piecewise Cuda Graph] rename, refactor and add more logging (#13675 )

2025-11-21 13:28:38 +08:00

tracing

Tiny log warn users when tracing is automatically disabled (#15889 )

2025-12-26 19:00:37 +08:00

utils

fix: warn once per env var key (#15846 )

2025-12-26 18:41:15 +08:00

weight_sync

[RL] Allow passing tensors of different dtypes for FlattenedTensorBucket (#13413 )

2025-11-19 17:24:48 +08:00

constants.py

Support releasing CUDA graph memory when paused (#7873 )

2025-10-28 14:40:50 +08:00

custom_op.py

Clean up the __init__ of TokenizerManager and DetokenizerManager (#15796 )

2025-12-25 10:30:27 -08:00

environ.py

Add a test case for crash dump (#15905 )

2025-12-26 19:39:28 -08:00

server_args_config_parser.py

[Lint] Add python/sglang to ruff F401 checks and remove unused imports in files (#11685 )

2025-10-17 16:49:46 -07:00

server_args.py

Add a test case for crash dump (#15905 )

2025-12-26 19:39:28 -08:00