sglang

batch_invariant_ops

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

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

checkpoint_engine

…

compilation

[Auto Sync] Update backend.py (20251130) (#14153 )

2025-11-30 22:15:02 -08:00

configs

Support nvidia/NVIDIA-Nemotron-Nano-12B-v2-VL-BF16 (and nvidia/C-RADIOv2-H) (#12277 )

2025-11-26 16:28:52 -07:00

connector

…

constrained

Support grammar + spec + reasoning (#14163 )

2025-11-30 21:19:57 +08:00

debug_utils

Super tiny add more info in dumper (#14145 )

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

disaggregation

[PD] Support json file configuration for Transfer Engine (#14059 )

2025-12-01 10:47:33 +08:00

distributed

[Piecewise] Use same global graph memory pool as the main cuda graph … (#14044 )

2025-11-30 23:04:10 -08:00

dllm

[feature] Initial block diffusion language model support (#12588 )

2025-11-26 17:57:54 +08:00

elastic_ep

…

entrypoints

Support profiling only prefill or decode without the other (#14182 )

2025-12-01 14:46:30 +08:00

eplb

Fix global scaling factor loading hang (#13484 )

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

function_call

Fix structural_tag tool call with null schema (#14006 )

2025-11-27 23:04:16 -08:00

grpc

Support numactl bind for CPU and memory before process starts (#14156 )

2025-11-30 17:00:33 +08:00

layers

Tiny fix transform_scale_ue8m0 wrong output in some scenarios (#14003 )

2025-12-01 14:45:27 +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 speculative decoding error when retracting (#14180 )

2025-12-01 15:30:13 +08:00

mem_cache

Fix LMCache unit test and init bug (#14005 )

2025-11-30 23:57:32 +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

Try to remove wrong logic about max total token in spec decoding (#14167 )

2025-12-01 15:29:58 +08:00

model_loader

Add adapter_model.safetensors to corruption validation for LoRA (#14022 )

2025-11-26 14:52:46 -08:00

models

[Feature] Enable PTPC FP8 for compressed tensors moe (aiter kernel) (#12181 )

2025-11-30 21:54:28 -08:00

multimodal

[VLM] Boost Memory Pool based CUDA IPC (#14123 )

2025-12-01 17:17:46 +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

…

speculative

Support grammar + spec + reasoning (#14163 )

2025-11-30 21:19:57 +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

[VLM] Boost Memory Pool based CUDA IPC (#14123 )

2025-12-01 17:17:46 +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

…

custom_op.py

…

environ.py

Super tiny allow millisecond precision in logging (#14183 )

2025-12-01 14:46:09 +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

…

server_args.py

[DeepSeekV3.2] Enable pure TP & Partial DP Attention (#13646 )

2025-11-30 15:59:23 -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

…