sglang

batch_invariant_ops

Add Batch‑Invariant RMSNorm (#12144 )

2025-10-28 21:05:57 -07:00

checkpoint_engine

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

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

compilation

[Fix] fix allreduce bug in Piecewise Graph (#12106 )

2025-10-26 21:15:48 +08:00

configs

Support Kimi Linear (#12469 )

2025-10-31 14:03:35 -07:00

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

[Feature] New structural tag support (#10691 )

2025-10-20 18:25:58 +08:00

debug_utils

feat: Add Non-intrusive Tensor Dumping for Model Inference (#10566 )

2025-10-31 12:04:48 +08:00

disaggregation

[HICache / PD]: Support offloading incremental KV cache in decode side. (#11966 )

2025-11-01 14:59:37 -07:00

distributed

Add prefix for torch symm mem (#12506 )

2025-11-02 11:23:05 -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

Try to allow NCCL cumem for multi node nvlink case (#11987 )

2025-10-31 12:48:25 -07:00

eplb

EPLB: prefer to use physical experts in the same gpu or node (#10874 )

2025-10-28 21:01:11 -07:00

function_call

[feat] Add SGLANG_TOOL_STRICT_LEVEL for tool-call behavior control (#12423 )

2025-11-01 13:15:02 -07:00

grpc

[bug] fix router installation to include additional dependency (#12348 )

2025-10-29 12:45:18 -07:00

layers

Add prefix for torch symm mem (#12506 )

2025-11-02 11:23:05 -08:00

lora

Feature/Add GET endpoint to query loaded LoRA adapters (#12229 )

2025-10-28 01:22:00 -07:00

managers

[FEAT] Shared mem pool based cuda ipc for multi-modal data transport (#11917 )

2025-11-02 16:46:37 +08:00

mem_cache

feat: Add FP4 (E2M1) KV Cache Support with Quantization Utilities for MLA (#10078 )

2025-11-01 22:24:58 -07:00

metrics

minor code sync (#12403 )

2025-10-30 12:49:31 -07:00

model_executor

Add prefix for torch symm mem (#12506 )

2025-11-02 11:23:05 -08:00

model_loader

Use Flashinfer TRT-LLM as Llama 4 compatible MoE backend (#11928 )

2025-10-28 10:39:43 -07:00

models

[FEAT] Shared mem pool based cuda ipc for multi-modal data transport (#11917 )

2025-11-02 16:46:37 +08:00

multimodal

[FEAT] Shared mem pool based cuda ipc for multi-modal data transport (#11917 )

2025-11-02 16:46:37 +08:00

multiplex

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

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

parser

[fix] added image token as prefix for deepseek-ocr (#12358 )

2025-10-30 16:37:03 +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

[Feature] Qwen3-Next & FLA: Support MTP topk>1; Up to 6% faster (#11133 )

2025-11-01 19:47:56 -07:00

tokenizer

move more files under srt/utils (#11285 )

2025-10-09 16:46:15 -07:00

tracing

[Feature] Sglang Tracing: Fine-Grained Tracking for Request Latency - Part 2 (#10804 )

2025-10-28 01:25:46 -07:00

utils

[FEAT] Shared mem pool based cuda ipc for multi-modal data transport (#11917 )

2025-11-02 16:46:37 +08:00

weight_sync

Tiny move files to utils folder (#11166 )

2025-10-03 22:40:06 +08:00

_custom_ops.py

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

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

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

[feat] Add SGLANG_TOOL_STRICT_LEVEL for tool-call behavior control (#12423 )

2025-11-01 13:15:02 -07:00

operations_strategy.py

[5/N] MoE Refactor: Update MoE parallelism arguments (#8658 )

2025-08-01 01:20:03 -07:00

operations.py

Add fp4 quantize before all-gather for Flashinfer cutlass MoE DP (max throughput) (#7667 )

2025-08-15 22:08:11 -07:00

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

[FEAT] Shared mem pool based cuda ipc for multi-modal data transport (#11917 )

2025-11-02 16:46:37 +08:00

single_batch_overlap.py

[hotfix] Incorrect CombineOverlapArgs in SBO (#12230 )

2025-10-28 01:23:06 -07:00

two_batch_overlap.py

[9/N] MoE Refactor: cleanup dispatcher interfaces (#11847 )

2025-10-20 10:11:46 -07:00

warmup.py

[Auto Sync] Update scheduler_profiler_mixin.py, rpd_utils.p... (20250916) (#10494 )

2025-09-16 17:02:20 -07:00