Search results

[Bug] [NPU] Diffusion models performance decreased after 17996 merged

Status: Open.

#20533 In sgl-project/sglang;

· OrangeRedeng opened on Mar 13, 2026

[Bug] [NPU] W4A4 MoE consumes more memory than expected

Status: Open.

#20532 In sgl-project/sglang;

· OrangeRedeng opened on Mar 13, 2026

[Bug] Shape mismatch in Indexer with GLM5 pp2

Status: Open.

#20529 In sgl-project/sglang;

· Zhangmj0621 opened on Mar 13, 2026

[Bug] Cuda graph with Torch Native LoRA Backend

Status: Open.

#20525 In sgl-project/sglang;

· vlserov opened on Mar 13, 2026

[diffuser] An Counterintuitive Test result on my custom diffusion model

Status: Open.

#20519 In sgl-project/sglang;

· JohnHerry opened on Mar 13, 2026

[Feature] Support IGW mode with PD disaggregation

Status: Open.

#20515 In sgl-project/sglang;

· purp1e-ace opened on Mar 13, 2026

[Tracking] SGLang CI/CD Test Coverage Improvements

cicontinue integration relatedcontinue integration related

nvidia

Status: Open.

#20514 In sgl-project/sglang;

· nvpohanh opened on Mar 13, 2026

[WIP][Tracking] Ensure Full E2E Tests for Key Models/Configs

cicontinue integration relatedcontinue integration related

nvidia

Status: Open.

#20513 In sgl-project/sglang;

· nvpohanh opened on Mar 13, 2026

[WIP][Tracking] Add Reduced-Layers E2E Tests

cicontinue integration relatedcontinue integration related

nvidia

Status: Open.

#20512 In sgl-project/sglang;

· nvpohanh opened on Mar 13, 2026

[WIP][Tracking] Add Kernel-Level Tests for External Backends

cicontinue integration relatedcontinue integration related

nvidia

Status: Open.

#20510 In sgl-project/sglang;

· nvpohanh opened on Mar 13, 2026

[Bug] Deterministic inference is not supported for Qwen3.5 next 400B FP8

Status: Open.

#20509 In sgl-project/sglang;

· Mikechen-0105 opened on Mar 13, 2026

[WIP][Tracking] Add Existing Hopper Tests to Blackwell CI

cicontinue integration relatedcontinue integration related

nvidia

Status: Open.

#20507 In sgl-project/sglang;

· nvpohanh opened on Mar 13, 2026