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

EP Support for Piecewise Cuda Graph (#14164 )

2025-12-20 01:59:27 +08:00

configs

Add MiDasheng Model Support (#15219 )

2025-12-19 21:41:59 +08:00

connector

…

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

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

2025-12-19 11:40:07 +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

vlm: refactor engine vlm params and support processor output as input (#14091 )

2025-12-20 18:31:24 +08:00

eplb

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

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

function_call

vlm: refactor engine vlm params and support processor output as input (#14091 )

2025-12-20 18:31:24 +08:00

grpc

[bug] fix grpc secheduler launcher breaking change (#15080 )

2025-12-13 10:36:49 -08:00

hardware_backend/npu

[NPU] [BUGFIX] [CRITICAL!] Fix NPU inference (torch_npu._npu_reshape_and_cache() crash) (#15484 )

2025-12-20 14:53:46 +08:00

layers

[FusedMoE] Fix fused w13 tp sharded weight loading (#15432 )

2025-12-21 13:59:36 +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

feat: Add limit-mm-data-per-request argument to server arguments (#15418 )

2025-12-20 10:44:47 -08:00

mem_cache

[EAGLE] Fix slow Triton compilation in EAGLE KV cache copy by chunking large num_locs_upper (#15111 )

2025-12-21 14:14:14 +08:00

metrics

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

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

model_executor

EP Support for Piecewise Cuda Graph (#14164 )

2025-12-20 01:59:27 +08:00

model_loader

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

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

models

Optimize Bailing-MoE with FlashInfer Fused All-Reduce (#15526 )

2025-12-20 17:34:04 -08:00

multimodal

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

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

multiplex

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

2025-11-01 00:40:01 +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

Tiny fix mimo model conflicts with main (#15483 )

2025-12-19 23:20:59 +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

[NPU]Fix for ipc handle with npu (#14138 )

2025-12-19 22:39:04 +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

…

environ.py

[amd] Add deterministic all-reduce kernel for AMD (ROCm) (#15340 )

2025-12-18 23:36:03 -08:00

server_args_config_parser.py

…

server_args.py

feat: Add limit-mm-data-per-request argument to server arguments (#15418 )

2025-12-20 10:44:47 -08:00