sglang

batch_invariant_ops

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

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

checkpoint_engine

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

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

compilation

Move unnecessary input_addr capture under debug mode flag for speed-up (#13690 )

2025-11-22 11:42:26 -08:00

configs

Feat/nemotron nano v3 support (#12690 )

2025-11-21 13:53:05 -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

fix xgrammar_backend crash with malformed inputs (#13752 )

2025-11-24 13:54:19 -08:00

debug_utils

Add script to create a model with fewer layers for debugging (#13284 )

2025-11-14 22:13:27 +08:00

disaggregation

[misc] Rename minilb install env & remove files & fix lint (#13831 )

2025-11-24 13:50:00 +08:00

distributed

fix uneven PP layer indices (#13282 )

2025-11-17 19:54:23 +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

[Fix] Fix uvloop get_event_loop() is not suitable for 0.22.x (#13612 )

2025-11-25 01:20:00 +08:00

eplb

Fix global scaling factor loading hang (#13484 )

2025-11-21 16:07:06 -08:00

function_call

[Tool Call] Steamline function arguments when tool_choice="auto" for deepseekv31_detector (#11589 )

2025-11-13 23:06:02 -08:00

grpc

[Fix] Fix uvloop get_event_loop() is not suitable for 0.22.x (#13612 )

2025-11-25 01:20:00 +08:00

layers

[Ascend] qwen optimization (#12078 )

2025-11-25 19:44:24 +08:00

lora

add LoRA warning if loading a preexisting LoRA adapter with a different name (#13822 )

2025-11-24 15:16:41 -08:00

managers

[Fix]: Further fix the buffer len of future map (#13916 )

2025-11-25 18:09:28 +08:00

mem_cache

[NPU] Fix NPU CI (#13834 )

2025-11-25 10:09:36 +08:00

metrics

CI: Kill zombie diffusion processes in CI & minor code style fix on rotary embedding fallback (#13637 )

2025-11-20 21:57:01 +08:00

model_executor

[Ascend] qwen optimization (#12078 )

2025-11-25 19:44:24 +08:00

model_loader

Fix model weights validation with automatic cache cleanup (#13729 )

2025-11-22 11:40:00 +08:00

models

[Ascend] qwen optimization (#12078 )

2025-11-25 19:44:24 +08:00

multimodal

[Performance] Replace preprocess_video logic from GLM multimodal processor with transformer impl for speed up (up to 27% faster) and addressing OOM (up to 50x improvements) (#13487 )

2025-11-24 18:17:13 -08:00

multiplex

[Feature] PD-Multiplexing Context and Scheduler, lazy import spatial. (#12275 )

2025-11-01 00:40:01 +08:00

parser

add Kimi k2 reasoning parser (#12702 )

2025-11-06 00:37:54 +08:00

sampling

[Bug] OOM (Out-of-Memory) errors for extreme testing scenarios (min_tokens=2) (#11757 )

2025-11-01 00:28:41 +08:00

speculative

[NPU] Fix NPU CI (#13834 )

2025-11-25 10:09:36 +08:00

tokenizer

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

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

tracing

[minor] remove debug code in python/sglang/srt/compilation/weak_ref_tensor_jit.py (#13235 )

2025-11-14 12:22:24 -08:00

utils

[Ascend] qwen optimization (#12078 )

2025-11-25 19:44:24 +08:00

weight_sync

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

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

_custom_ops.py

move all get_stream in sgl_kernel to c++ to reduce the launch overhead (#12521 )

2025-11-02 13:15:05 -08:00

constants.py

Support releasing CUDA graph memory when paused (#7873 )

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

custom_op.py

…

environ.py

Fix: Safe RoPE Cache Expansion to Prevent Position-ID Out-of-Bounds in EAGLE + Long-Sequence Workloads (#11871 )

2025-11-25 01:19:06 +08:00

operations_strategy.py

…

operations.py

Register allgather/reducescatter buffers with symm memory (#12572 )

2025-11-04 17:11:36 -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

[Ascend] qwen optimization (#12078 )

2025-11-25 19:44:24 +08:00

single_batch_overlap.py

[11/N] MoE Refactor: Simplifying SBO Implementation with Dispatcher Hooks (#13327 )

2025-11-21 01:11:37 -08:00

two_batch_overlap.py

[11/N] MoE Refactor: Simplifying SBO Implementation with Dispatcher Hooks (#13327 )

2025-11-21 01:11:37 -08:00

warmup.py

…