Search results

[Feature] Support Moore Threads (MUSA) GPU

Status: Open.

#16565 In sgl-project/sglang;

· yeahdongcn opened on Jan 6, 2026

[Bug] accuracy not tested in test_decode_attention

Status: Open.

#16554 In sgl-project/sglang;

· polisettyvarma opened on Jan 6, 2026

[Bug] Failed to Infer Qwen2.5-7B

Status: Open.

#16553 In sgl-project/sglang;

· yg9538 opened on Jan 6, 2026

Support Qwen3-Next-80B-A3B-Instruct linear attention modules (in_proj_qkvz/in_proj_ba) for LoRA

Status: Open.

#16550 In sgl-project/sglang;

· 1277331747 opened on Jan 6, 2026

[RFC] SGLang-Omni Design

Status: Open.

#16546 In sgl-project/sglang;

· FrankLeeeee opened on Jan 6, 2026

[Bug] [model-gateway] External workers fail with "Unknown parameter" in IGW mode

Status: Open.

#16544 In sgl-project/sglang;

· zhaowenzi opened on Jan 6, 2026

[Bug] Structured outputs not working with spec v2 PD and spec v1 crashes - DeepseekV3.1 / mimo-v2-flash

Status: Open.

#16541 In sgl-project/sglang;

· Tiktus opened on Jan 6, 2026

[Bug] DecodingStage VRAM usage surges dramatically

Status: Open.

#16530 In sgl-project/sglang;

· carloszhang999 opened on Jan 6, 2026

[Bug] on NPU,the server crashed when add --enable-piecewise-cuda-graph

Status: Open.

#16526 In sgl-project/sglang;

· fourierr opened on Jan 6, 2026

[Bug] glmasr: cannot import name 'GlmAsrConfig' from 'transformers'

Status: Open.

#16518 In sgl-project/sglang;

· universorum opened on Jan 5, 2026

[Bug] dsv3 continue_fina_message error

Status: Open.

#16501 In sgl-project/sglang;

· whybeyoung opened on Jan 5, 2026

[Bug] NPU Runtime Error in graph_runner.replay when batchsize changed during running benchmarking Qwen3+Eagle3

Status: Open.

#16498 In sgl-project/sglang;

· M0reDr1nk opened on Jan 5, 2026