Problemsby GlyphTrouble with vLLM speculative
3 posts from 3 people3d activeFading, +0 this week
Posts per day
The posts behind it
[Bug]: Hybrid GDN models (Qwen3.5) with n-gram speculative decoding: greedy outputs go wrong when drafts are partly accepted; the GDN state misses the last accepted token's update
...later steps read slots. So the output reveals which writes the state contains. For every failing request, replaying the oracle shows the state was **missing the write of a token that had been accepted earlier**, most often the token accepted in the step just before the first...
[Bug]: Model Runner V2 hangs during speculative decoding warmup with pipeline parallelism
### Your current environment The output of python collect_env.py ### 🐛 Describe the bug **Description** Code version: based on `main` commit [`[de7eeaa87bc19ecf088472d5f4471cdc9a5ab995](https://github.com/vllm-project/vllm/commit/de7eeaa87bc19ecf088472d5f4471cdc9a5ab995)`](https...
Support Qwen3.x MTP (Multi-Token Prediction) for speculative decoding
## Summary Following analysis of 1Cat-vLLM's MTP-based speculative decoding for Qwen3.6-35B-A3B, this issue proposes adding MTP (Multi-Token Prediction) speculative decoding support f...[truncated]
Companies and products named
About this problem
Evidence
3 posts from 3 people in 2 places, about 1 a week over 50 days. Mostly on vllm-project/vllm, ggml-org/llama.cpp. Tools named alongside: NVIDIA, llama.cpp, GitHub, Oracle.
Frustration Frustration 0 of 3· Seen on GitHub issues
How it was grouped
Posts about vLLM with no known issue, sharing the word "speculative". First post Aug 30, 2026, latest Oct 18, 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.
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