Search results

Request: Backend Support for Domestic GPU Chips

Status: Open.

#47 In stepfun-ai/StepMesh;

· lipracer opened on Dec 11, 2025

[Issue] RDMA PushRequest protocol mismatch causes ibv_post_send failure and data_num != 3 assertion on non-GDR (Ascend NPU) environment

Status: Open.

#46 In stepfun-ai/StepMesh;

· Dong-Jiahuan opened on Dec 11, 2025

issues of single node example

Status: Open.

#44 In stepfun-ai/StepMesh;

· menggerSherry opened on Nov 15, 2025

Problem trying to get vLLM + Step3 + AFD running

Status: Open.

#43 In stepfun-ai/StepMesh;

· orionorin opened on Nov 11, 2025

StepMesh performance issue

Status: Open.

#42 In stepfun-ai/StepMesh;

· JF-D opened on Nov 7, 2025

Tensor size check failed in unit test

Status: Open.

#39 In stepfun-ai/StepMesh;

· JF-D opened on Sep 28, 2025

[Question] Why is the communication between Worker and Server Restricted to GPU with same local id

enhancementNew feature or requestNew feature or request

Status: Open.

#31 In stepfun-ai/StepMesh;

· Zhangmj0621 opened on Aug 28, 2025

[Bug] Get crashed when change batchsize in bmk_comm_latency_multiserver.py

Status: Open.

#28 In stepfun-ai/StepMesh;

· Zhangmj0621 opened on Aug 22, 2025

[BUG] Crashed when scaling out to 4 workers 4 servers.

Status: Open.

#27 In stepfun-ai/StepMesh;

· Zhangmj0621 opened on Aug 22, 2025

How you implement 3-stage pipeline communication with StepMesh?

documentationImprovements or additions to documentationImprovements or additions to documentation

Status: Open.

#18 In stepfun-ai/StepMesh;

· Huixxi opened on Aug 12, 2025