Search results

Bug: vllm-plugin-FL error on nvidia (Qwen/Qwen3.5-0.8B)

Status: Open.

#157 In flagos-ai/vllm-plugin-FL;

· buzhengjing opened on May 8, 2026

Bug: vllm-plugin-FL error: dispatch, vllm_fl on nvidia (moonshotai/Kimi-K2.6)

Status: Open.

#153 In flagos-ai/vllm-plugin-FL;

· buzhengjing opened on Apr 28, 2026

[Bug]:

bugSomething isn't workingSomething isn't working

Status: Open.

#142 In flagos-ai/vllm-plugin-FL;

· huangyiqun opened on Apr 13, 2026

ValueError: Vendor 'hygon'/'amd' not found in VENDOR_DEVICE_MAP on Hygon DCU hardware

Status: Open.

#126 In flagos-ai/vllm-plugin-FL;

· JosephNew opened on Apr 9, 2026

[Bug]: support_static_graph_mode does not include "hygon" vendor, preventing CUDA graph on Hygon DCU

bugSomething isn't workingSomething isn't working

Status: Open.

#122 In flagos-ai/vllm-plugin-FL;

· JiaryCoder opened on Apr 8, 2026

[docs][Ascend] LLM initialization fails with "Device string must not be empty" when using offline.py

docs

Status: Open.

#120 In flagos-ai/vllm-plugin-FL;

· gbdjxgp opened on Apr 7, 2026

[Bug]: MACA (MetaX) platform patches missing from repository — Triton kernel shared memory OOM on MetaX GPUs

bugSomething isn't workingSomething isn't working

Status: Open.

#93 In flagos-ai/vllm-plugin-FL;

· buzhengjing opened on Mar 17, 2026

[Bug] Torch conflict when using vllm-plugin-FL on BI-V150

Status: Open.

#27 In flagos-ai/vllm-plugin-FL;

· shh2000 opened on Jan 28, 2026

ModuleNotFoundError: No module named 'flag_gems.fused.FLA'

Status: Open.

#23 In flagos-ai/vllm-plugin-FL;

· CaiJichang212 opened on Jan 23, 2026

[Bug] plugin conflict when using vllm-plugin-fl on C550

Status: Open.

#6 In flagos-ai/vllm-plugin-FL;

· shh2000 opened on Dec 26, 2025

[Bug and Feature Request] Request for decoupling FlagGems/FlagCX dependencies + README fixes + reproducible inference environment (including non-NVIDIA hardware)

Status: Open.

#5 In flagos-ai/vllm-plugin-FL;

· shh2000 opened on Dec 26, 2025