Problemsby GlyphTrouble with vLLM size
3 posts from 3 people3d activeSteady, +1 this week
Posts per day
The posts behind it
[Bug]: MoE w2_bias is added tp_size times under tensor parallelism
### Your current environment vLLM 0.32.0 (release CPU wheel), PyTorch 2.13.0+cpu, fp32, gloo between processes. The affected code is unchanged on main (40837c5d0e). ### 🐛 Describe the bug `RoutedExperts.weight_loader` copies the full checkpoint `w2_bias` onto every TP rank. Its...
[Bug]: EPLB counts load for the wrong expert when step_interval < window_size
### Your current environment Any EPLB setup with eplb_config step_interval smaller than window_size. ### Describe the bug EPLB records load per physical slot. When step_interval < window_size only step_interval rows of the window are refreshed between two rearranges. The other...
[Bug]: `vllm preload` daemon never matches a dense (non-MoE) engine with `--data-parallel-size > 1` (WeightCacheKey `dp_size` mismatch)
...holding its own copy of the weights on every GPU. - With `"fallback": false`, the engine fails to start. The preload docs list data parallelism as supported ([`docs/features/preload.md`...
Companies and products named
About this problem
Evidence
3 posts from 3 people in 1 place, about 1 a week over 29 days. Mostly on vllm-project/vllm. Tools named alongside: GitHub, PyTorch, Weights & Biases.
Frustration Frustration 0 of 3· Seen on GitHub issues
How it was grouped
Posts about vLLM with no known issue, sharing the word "size". First post Oct 6, 2026, latest Nov 3, 2026. Corroborated: 3 or more posts from 2 or more people or places. 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.
vLLM runs out of memory
3 posts from 3 people on GitHub issues.
vLLM fails to return valid structured output
3 posts from 3 people on GitHub issues.
History