Problemsby GlyphNVIDIA runs out of context on real work
2 posts from 2 people2d activeSteady, +1 this weekunverified
Posts per day
The posts behind it
Proposal: CUDA GQA flash attention for q8_0 K / q4_0 V
**Status & Context:** Discussion/coordination with ongoing upstream work; not proposed as an independent PR. Relevant upstream efforts include [PR #30792](https://github.com/ggml-org/llama.cpp/pull/30792) and [issue #30790](https://github.com/ggml-org/llama.cpp/issues/30790)...
Flash attention (CUDA/HIP): garbage output at long context on gfx1201 (RDNA4) - immediate EOS, head_dim 256
With flash attention enabled, long-context prefill produces corrupted logits on gfx1201 (AMD Radeon AI R9700, RDNA4) for a model with head_dim 256 (Qwen3.8-27B). The first generated token is EOS with high confidence (generation stops immediately, no decode step runs), so the...
Companies and products named
About this problem
Evidence
2 posts from 2 people in 1 place, about 1 a week over 18 days. Mostly on ggml-org/llama.cpp. Tools named alongside: llama.cpp, GitHub, Benchmark.
Frustration Frustration 3 of 3· Seen on GitHub issues
How it was grouped
Posts that state a pain and match the "context-window" rule about NVIDIA. First post Oct 20, 2026, latest Nov 6, 2026. Unverified: fewer than 3 posts, or one person or place only. Method
The same problem elsewhere
Context windows run out on real workloads
20 posts from 20 people on Hacker News, Reddit, GitHub issues.
Claude Code runs out of context on real work
12 posts from 12 people on GitHub issues, Reddit, Hacker News.
Claude runs out of context on real work
10 posts from 8 people on GitHub issues, Reddit, Hacker News.
OpenAI runs out of context on real work
8 posts from 8 people on GitHub issues, Community forums.
ChatGPT runs out of context on real work
8 posts from 8 people on GitHub issues, Reddit, Hacker News.
Ollama runs out of context on real work
7 posts from 7 people on GitHub issues, Reddit.
Codex runs out of context on real work
6 posts from 6 people on GitHub issues, Reddit, Hacker News.
Cursor runs out of context on real work
5 posts from 5 people on Community forums, Reddit, Hacker News.
vLLM runs out of context on real work
5 posts from 5 people on GitHub issues.
Mistral runs out of context on real work
4 posts from 4 people on GitHub issues.
Anthropic runs out of context on real work
4 posts from 4 people on GitHub issues.
llama.cpp runs out of context on real work
4 posts from 4 people on GitHub issues.
Other NVIDIA problems
Running NVIDIA locally hits memory and speed limits
16 posts from 16 people on GitHub issues.
NVIDIA users keep filing bugs
Top post: "`clef-flash` decision model always fails on `/v1/systemone`, Clef: non-finite logit (CUDA) / Clef: cannot", one of 41 posts from 38 people on GitHub issues, Reddit, Community forums.
Trouble with NVIDIA windows
Top post: "Windows CUDA prebuilt binaries stuck at b11828 while release tags keep advancing (b11846..b11852)", one of 3 posts from 3 people on GitHub issues.
NVIDIA GPU access and drivers are painful
12 posts from 12 people on GitHub issues, Mastodon, Reddit.
NVIDIA vulnerabilities are hard to triage and patch
77 posts from 77 people on Glyph Ibis.
NVIDIA upgrades break working setups
11 posts from 11 people on GitHub issues.
Trouble with NVIDIA account
Top post: "NVIDIA Careers Workday Account Locked - Unable to Complete Employee Referral Application", one of 9 posts from 8 people on Community forums.
NVIDIA lock-in makes switching costly
5 posts from 5 people on GitHub issues, Hacker News, Bluesky.
NVIDIA runs out of memory
5 posts from 5 people on GitHub issues.
Trouble with NVIDIA slower
Top post: "CUDA training loop runs ~50x slower while a DataLoader iterates in a background thread, since 2.10.0", one of 4 posts from 4 people on GitHub issues.
Trouble with NVIDIA process
Top post: "`HuggingFaceEmbeddings(multi_process=True)` starts CPU workers on non-CUDA accelerators", one of 3 posts from 3 people on GitHub issues.
Trouble with NVIDIA phone
Top post: "Unable to receive OTP for Vietnam (+84) phone verification on build.nvidia.com", one of 3 posts from 3 people on Community forums.
History