Problemsby GlyphTrouble with vLLM attention
2 posts from 2 people2d activeFading, +0 this weekunverified
Posts per day
The posts behind it
[FA4][SM100] Support head_dim=256 with seqused_k for paged varlen attention
## Summary ### Agent Notes FA4's SM100 `head_dim=256` forward specialization rejects `seqused_q` / `seqused_k`. This blocks Qwen3.5-9B inference through PyTorch's `torch.nn.attention.varlen.varlen_attn_out` in vLLM, where causal paged KV attention supplies `block_table` and...
[Bug][CPU] GLM-5.x (glm_moe_dsa) cannot run on CPU: sparse attention forced by config, no way to disable
## Environment - vLLM version: v0.28.0 - Platform: Apple Silicon (M3), macOS 15.7.4, CPU backend - HF transformers: 5.16.1 - Model: `zai-org/GLM-5.2` (`glm_moe_dsa` / `GlmMoeDsaForCausalLM`) ## Bug Any GLM-5.x model (DSA / sparse-attention architecture) fails during engine...
Companies and products named
About this problem
Evidence
2 posts from 2 people in 2 places, about 2 a week over 6 days. Mostly on vllm-project/vllm, pytorch/pytorch. Tools named alongside: PyTorch, NVIDIA.
Frustration Frustration 3 of 3· Seen on GitHub issues
How it was grouped
Posts about vLLM with no known issue, sharing the word "attention". First post Aug 27, 2026, latest Sep 1, 2026. Unverified: fewer than 3 posts, or one person or place only. Method
Other vLLM problems
Running vLLM locally hits memory and speed limits
21 posts from 20 people on GitHub issues, Reddit.
Trouble with vLLM draft
Top post: "Draft model construction resets data_parallel_index to 0 (load_draft_model / create_draft_parallel_config),", one of 3 posts from 3 people on GitHub issues.
vLLM users keep filing bugs
Top post: "Qwen3.8-flash-next 0% MTP acceptance rate in disaggregated PD serving", one of 34 posts from 31 people on GitHub issues, Reddit.
vLLM upgrades break working setups
17 posts from 17 people on GitHub issues.
vLLM vulnerabilities are hard to triage and patch
9 posts from 9 people on Glyph Ibis.
vLLM runs out of context on real work
5 posts from 5 people on GitHub issues.
vLLM GPU access and drivers are painful
5 posts from 3 people on GitHub issues.
Fine-tuning vLLM is costly and unpredictable
4 posts from 4 people on GitHub issues.
Trouble with vLLM cache
Top post: "AOT compile cache never loads for Qwen3.5 models: Kernel index 1 not found in id_to_kernel", one of 4 posts from 4 people on GitHub issues.
Trouble with vLLM kernel
Top post: "ROCM_AITER_MLA can select an aiter MLA decode kernel that silently returns uncorrelated output (gqa_ratio=16,", one of 3 posts from 3 people on GitHub issues.
Trouble with vLLM size
Top post: "EPLB counts load for the wrong expert when step_interval < window_size", one of 3 posts from 3 people on GitHub issues.
vLLM runs out of memory
3 posts from 3 people on GitHub issues.
History