sglang

arg_groups

[1/N] elastic-ep: Add runtime EP scale-up (#30164 )

2026-07-16 15:53:44 -07:00

batch_invariant_ops

[NPU]adaptation to support deterministic inference (#21197 )

2026-06-09 09:22:09 +08:00

batch_overlap

Remove dead ScheduleBatch fields and avoid inplace seq_lens bump (#30669 )

2026-07-15 14:23:31 +08:00

checkpoint_engine

…

compilation

[Intel GPU] xpu_piecewise: fall back to eager when PCG capture stream is unset (#30235 )

2026-07-09 10:21:43 +08:00

configs

Remove QServe and FBGEMM FP8 quantization (#31109 )

2026-07-17 17:10:34 -07:00

connector

[connector] Add Azure Blob Storage connector (az:// and *.blob.core.windows.net URLs) (#23995 )

2026-05-11 14:20:22 -07:00

constrained

Upgrade llguidance to 1.7.6 (#31484 )

2026-07-17 16:31:44 -07:00

debug_utils

[Fix] Update PR #25015 revert for fused topk=1 draft postprocess (#31519 )

2026-07-16 21:51:16 -07:00

disaggregation

Fix Heartbeat Checker in KV Manager Disaggregation (#31584 )

2026-07-18 01:21:50 +08:00

distributed

[Kernel] Rewrite JIT custom all-reduce (v2) with a decoupled kernel/storage design (#31049 )

2026-07-17 18:37:22 +08:00

dllm

Avoid implicit running_batch access in dllm and pdmux scheduling (#30676 )

2026-07-15 14:33:00 +08:00

elastic_ep

[1/N] elastic-ep: Add runtime EP scale-up (#30164 )

2026-07-16 15:53:44 -07:00

entrypoints

Revert "Bump FlashInfer to 0.6.15 and revert regressions" (#31625 )

2026-07-17 16:46:33 -07:00

eplb

[1/N] elastic-ep: Add runtime EP scale-up (#30164 )

2026-07-16 15:53:44 -07:00

function_call

[minimax-m3] Split 4/4: model + VL + glue + function-call + fp8 quant + generic infra (#28715 )

2026-07-11 11:11:06 +08:00

grpc

…

hardware_backend

[Bugfix][NPU] Fix/Refactor routed scaling factor application in MoE routing (#31449 )

2026-07-18 10:59:13 +08:00

kv_canary

[Spec] Extract stateless draft prepare helpers into eagle_worker_common (#31257 )

2026-07-15 10:54:31 -07:00

layers

[Spec] DFlash: remove per-step host syncs so the CPU runs a full step ahead (spec-v2 overlap) (#31468 )

2026-07-17 23:22:07 -07:00

lora

refactor logprob processor layer (#20071 )

2026-07-17 16:28:10 -07:00

managers

fix(rpc) Synchronize RPC requests only within the TP group. (#25213 )

2026-07-18 07:44:51 +00:00

mem_cache

[Feature] Add FP4 KV Cache Design and support SM120 GPUs (#21601 )

2026-07-17 14:49:43 -07:00

model_executor

[CP] Migrate MLA prefill CP (DeepSeek V3) to CP-v2 zigzag strategy (#31619 )

2026-07-17 22:37:53 -07:00

model_loader

Extract leaf helpers out of ModelRunner into utility modules (#31146 )

2026-07-14 15:52:03 +08:00

models

[CP] Migrate MLA prefill CP (DeepSeek V3) to CP-v2 zigzag strategy (#31619 )

2026-07-17 22:37:53 -07:00

multimodal

optimization: shard kimi dp image feature transport and misc optimizations (#31227 )

2026-07-16 20:31:42 +08:00

multiplex

Avoid implicit running_batch access in dllm and pdmux scheduling (#30676 )

2026-07-15 14:33:00 +08:00

observability

[Bugfix][HiCache] measure load-back duration with CUDA events (#26411 )

2026-07-16 00:11:04 -07:00

parser

more fixes for Nemotron 3 parser for tool call and force nonempty content (#30533 )

2026-07-14 06:50:27 +08:00

platforms

[plugin] oot torch profiler activity support (#31580 )

2026-07-17 22:20:54 -07:00

plugins

Multi platform Plugin (#21388 )

2026-04-19 17:23:51 -07:00

ray

[refactor] Move model-capability adjustments into the resolution pipeline (#30299 )

2026-07-07 21:26:55 -07:00

sampling

Return top-p/top-k sampling mask/nucleas (#27408 )

2026-07-13 17:58:51 -05:00

session

Fix abusing presence of req.req_pool_idx to indicate the presence of req.kv resources (#29430 )

2026-07-15 14:48:39 +08:00

speculative

[Spec] DFlash: remove per-step host syncs so the CPU runs a full step ahead (spec-v2 overlap) (#31468 )

2026-07-17 23:22:07 -07:00

state_capturer

Absorb capturer setup and extract the shared-mooncake gate (#31160 )

2026-07-14 16:00:51 +08:00

tokenizer

Fix TiktokenTokenizer missing num_special_tokens_to_add (#30643 )

2026-07-10 12:37:30 +00:00

utils

[plugin] oot torch profiler activity support (#31580 )

2026-07-17 22:20:54 -07:00

weight_sync

Introduce WeightUpdater and WeightExporter components (#31148 )

2026-07-14 15:53:33 +08:00

constants.py

[AMD][MXFP4] Online MXFP4 quantization 1/N - dense and MOE models w. original BF16 weight (#18005 )

2026-06-03 12:55:24 -07:00

environ.py

[Spec] DFlash: remove per-step host syncs so the CPU runs a full step ahead (spec-v2 overlap) (#31468 )

2026-07-17 23:22:07 -07:00

runtime_context.py

[1/N] elastic-ep: Add runtime EP scale-up (#30164 )

2026-07-16 15:53:44 -07:00

server_args_config_parser.py

fix(server): serialize nested dict config values as JSON (#28094 )

2026-06-12 19:27:31 -07:00

server_args.py

Remove QServe and FBGEMM FP8 quantization (#31109 )

2026-07-17 17:10:34 -07:00