sglang

batch_invariant_ops

[Auto Sync] Update batch_invariant_ops.py (20260221) (#19098 )

2026-02-20 18:27:31 -08:00

batch_overlap

fix retrive -> retrieve typo (#23503 )

2026-04-22 16:35:04 -07:00

checkpoint_engine

update pre-commit config (#18860 )

2026-02-16 00:18:31 +08:00

compilation

Multi platform Plugin (#21388 )

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

configs

[Feature] Xiaomi MiMo-V2.5-Pro day0 support (#23808 )

2026-04-28 11:43:29 +08:00

connector

Remove sync points in mamba cache + prefill cudagraph plumbing for DP (#19639 )

2026-03-02 15:03:42 +08:00

constrained

[Fix] eagle/eagle3 speculative decoding conflicts with xgrammar in NPU (#20989 )

2026-04-15 23:34:23 -07:00

debug_utils

Add e2e demo test in dump comparator (#21031 )

2026-03-20 22:41:01 +08:00

disaggregation

tokenizer: Add fastokens support (#23753 )

2026-04-28 11:43:10 -07:00

distributed

[6/N] (Elastic EP) Recover failed ranks (#15771 )

2026-04-28 00:44:26 -07:00

dllm

Decouple scheduler log printing from metrics collection (#20107 )

2026-03-07 22:09:10 -08:00

elastic_ep

[6/N] (Elastic EP) Recover failed ranks (#15771 )

2026-04-28 00:44:26 -07:00

entrypoints

Fix multimodal /v1/embeddings Jinja chat template handling (#20835 )

2026-04-28 13:05:45 -07:00

eplb

[6/N] (Elastic EP) Recover failed ranks (#15771 )

2026-04-28 00:44:26 -07:00

function_call

support Hy3 preview (#23533 )

2026-04-24 12:03:24 -07:00

grpc

[gRPC] Extract gRPC servicer into standalone package (#20478 )

2026-03-13 09:13:29 -07:00

hardware_backend

【NPU】【bugfix】accuracy fix when enable both nsa cp and prefixcache (#23268 )

2026-04-28 09:08:28 +08:00

layers

【NPU】【bugfix】accuracy fix when enable both nsa cp and prefixcache (#23268 )

2026-04-28 09:08:28 +08:00

lora

fix(lora): don't assert on non-LoRA lm_head adapter weights (#23433 )

2026-04-26 12:10:07 -07:00

managers

[Spec] Fix spec_accept_rate and unify accept/draft naming (#23530 )

2026-04-28 14:40:04 -07:00

mem_cache

[HiCache][SPEC] fix: empty key after page alignment in match_prefix (#23387 )

2026-04-28 14:06:55 -07:00

model_executor

[6/N] (Elastic EP) Recover failed ranks (#15771 )

2026-04-28 00:44:26 -07:00

model_loader

[AMD] Fix Kimi-K2.6 Quark MXFP4 loading prefix and packed module mapping (#23408 )

2026-04-26 23:56:15 -07:00

models

[Fix] NVFP4 qwen3.5 quant error fix by add packed_modules_mapping (#23471 )

2026-04-28 13:36:09 -07:00

multimodal

[VLM] Fix Kimi-K2.5 CPU path: rename grid_thws -> image_grid_thw (#23501 )

2026-04-27 20:34:28 +00:00

multiplex

[mem] Flatten memory checkers into composable per-pool invariant checks (#22562 )

2026-04-11 02:56:22 -07:00

observability

[Spec] Fix spec_accept_rate and unify accept/draft naming (#23530 )

2026-04-28 14:40:04 -07:00

parser

support Hy3 preview (#23533 )

2026-04-24 12:03:24 -07:00

platforms

Clean up noisy startup warnings from third-party deps (#23669 )

2026-04-27 03:10:46 -07:00

plugins

Multi platform Plugin (#21388 )

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

ray

[Ray] Bind scheduler actors to GPU-local NUMA node (#22989 )

2026-04-16 14:52:15 -07:00

sampling

Fix /generate endpoint crash when sampling params contain null values (#23401 )

2026-04-22 09:56:10 -07:00

session

[bug] Fix cache salt and extra keys for prefix cache isolation (#23300 )

2026-04-21 13:53:24 -07:00

speculative

[Spec] Fix spec_accept_rate and unify accept/draft naming (#23530 )

2026-04-28 14:40:04 -07:00

tokenizer

[Piecewise Cuda Graph] rename, refactor and add more logging (#13675 )

2025-11-21 13:28:38 +08:00

utils

tokenizer: Add fastokens support (#23753 )

2026-04-28 11:43:10 -07:00

weight_sync

[RL] Allow passing tensors of different dtypes for FlattenedTensorBucket (#13413 )

2025-11-19 17:24:48 +08:00

constants.py

Clean up detokenizer and remove dead multimodal_gen code (#21588 )

2026-03-28 21:44:40 -07:00

environ.py

[NPU] Support MTP for Qwen3.5 (#20918 )

2026-04-27 10:44:17 +08:00

server_args_config_parser.py

[ConfigArgumentMerger] Improve ConfigArgumentMerger compatibility with external callers (#17051 )

2026-01-16 15:32:42 +08:00

server_args.py

tokenizer: Add fastokens support (#23753 )

2026-04-28 11:43:10 -07:00