sglang

batch_invariant_ops

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

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

batch_overlap

Single Batch Overlap for MoE Models (#9660 )

2025-12-03 10:07:42 -08:00

checkpoint_engine

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

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

compilation

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

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

configs

[Auto Sync] Rename is_hybrid to is_hybrid_swa (#14252 )

2025-12-01 23:24:24 -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

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

Support PP x PD decode with nixl backend (#14392 )

2025-12-04 12:22:17 +08:00

distributed

Move custom_ops under layers; move _custom_ops.py → custom_all_reduce_ops.py (#14326 )

2025-12-03 10:33:37 -08:00

dllm

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

2025-11-26 17:57:54 +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

Clean up imports and move files (#14317 )

2025-12-02 16:31:54 -08:00

eplb

Fix global scaling factor loading hang (#13484 )

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

function_call

feat: DeepSeek new v3.2 encoding (#14249 )

2025-12-02 11:41:05 -08:00

grpc

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

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

layers

Move custom_ops under layers; move _custom_ops.py → custom_all_reduce_ops.py (#14326 )

2025-12-03 10:33:37 -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

[bugfix] Fix prefill tbo disabled when --deepep-mode=auto (#14333 )

2025-12-03 01:20:33 -08:00

mem_cache

[Performance] Optimize NSA Indexer K/S Buffer Access with Fused Triton Kernels (#13812 )

2025-12-02 18:53:06 -08:00

metrics

[Auto Sync] Add max_total_num_tokens metric: Update scheduler_metrics_mixin.py, collector.py (20251202) (#14256 )

2025-12-01 16:34:34 -08:00

model_executor

Clean up imports and move files (#14317 )

2025-12-02 16:31:54 -08:00

model_loader

Fix validation to detect missing model files before loading (#14253 )

2025-12-03 11:36:07 -08:00

models

Single Batch Overlap for MoE Models (#9660 )

2025-12-03 10:07:42 -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

[Minor] update docs (#14212 )

2025-12-01 02:33:58 -08:00

speculative

[spec-overlap] bugfix for pd disaggregation and npu (#14088 )

2025-12-01 22:58:20 +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

Revert "Revert "enable csgmv automatically on cuda"" (#14277 )

2025-12-03 13:12:30 -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

enable llama3.1-8B on xpu (#9434 )

2025-09-07 22:34:20 -07:00

environ.py

[bug fix] use npu phy id in container env (#14266 )

2025-12-03 11:33:43 +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

Support fp4 fp8 non gated moe (#13794 )

2025-12-01 15:26:28 -08:00