Yuan Luoandluoyuan.luo 3912ee4991 [VLM] feat: support chunked vit attention (#14907) ...

Co-authored-by: luoyuan.luo <luoyuan.luo@antgroup.com>

2025-12-15 12:11:02 +08:00

..

batch_invariant_ops

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

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

batch_overlap

[bugfix] fix TBO crashes when attn_tp_size > 1 (#13730)

2025-12-11 18:18:40 -08:00

checkpoint_engine

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

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

compilation

[Ascend]Support of piecewise graph compilation for prefill on NPU (#12287)

2025-12-11 21:10:07 +08:00

configs

feat: support EPD disaggregation (#12263)

2025-12-14 22:30:08 +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

[refactor] Update reasoning parameter to require_reasoning (#14922)

2025-12-11 16:55:59 -08:00

debug_utils

Super tiny add more info in dumper (#14145)

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

disaggregation

[Qwen3-next] support mamba radix cache for overlap scheduler (#14792)

2025-12-14 18:54:16 -08:00

distributed

[NPU] adapt dsv3.2 nsa prefill context parallel (#14541)

2025-12-11 20:12:59 +08:00

dllm

[DLLM] feat: Add threshold based parallel decoding support (#14412)

2025-12-07 18:25:33 +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

Add NanoV3 reasoning parser support (#15113)

2025-12-14 12:09:44 -08:00

eplb

[NPU][1/N] NPU basic functions refactor and new modelslim quant type (#13359)

2025-12-04 16:15:31 +08:00

function_call

Fix GLM-4.6 tool calls don't support streaming output for arguments i… (#13989)

2025-12-13 00:37:18 -08:00

grpc

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

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

hardware_backend/npu

[NPU] perf update with kvcache nz & w4a8 quant (#14423)

2025-12-13 17:39:55 +08:00

layers

[Qwen3-next] support mamba radix cache for overlap scheduler (#14792)

2025-12-14 18:54:16 -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

[VLM] feat: support chunked vit attention (#14907)

2025-12-15 12:11:02 +08:00

mem_cache

[Qwen3-next] support mamba radix cache for overlap scheduler (#14792)

2025-12-14 18:54:16 -08:00

metrics

Add sglang:decode_sum_seq_lens metric (#15066)

2025-12-14 08:42:45 +08:00

model_executor

[Qwen3-next] support mamba radix cache for overlap scheduler (#14792)

2025-12-14 18:54:16 -08:00

model_loader

Revert several PRs (#14958)

2025-12-12 11:25:12 -08:00

models

[VLM] feat: support chunked vit attention (#14907)

2025-12-15 12:11:02 +08:00

multimodal

feat: support EPD disaggregation (#12263)

2025-12-14 22:30:08 +08:00

multiplex

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

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

parser

Add NanoV3 reasoning parser support (#15113)

2025-12-14 12:09:44 -08:00

sampling

[Glm46v] Bug fix for accuracy drop and unable to launch server (#14585)

2025-12-07 23:45:02 -08:00

speculative

[Qwen3-next] support mamba radix cache for overlap scheduler (#14792)

2025-12-14 18:54:16 -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

Enable TRT AllReduce Fusion by default (#14764)

2025-12-14 20:01:49 -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

Introduce native kv cache move (#15108)

2025-12-14 17:23:28 +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

Enable TRT AllReduce Fusion by default (#14764)

2025-12-14 20:01:49 -08:00