llmfit · Issues· 69 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #1039
Weekly model refresh fails since 2026-08-31: HF 429 rate limiting degrades the scrape and trips the #963 guard
bugAutomationUpdated Sep 16, 2026 - #301
Request to add support for NPU and Hybrid models
enhancementFeature RequestHardware CompatibilityUpdated Sep 15, 2026 - #176
[Feature Request] Sort by release date
Feature RequestHardware CompatibilitySearch FunctionalityUpdated Sep 15, 2026 - #1040
[Bug]: `llmfit bench` silently substitutes an unrelated model when the requested tag isn't installed, instead of erroring
bugUpdated Sep 15, 2026 - #1045
[Bug]: pre-quantized AWQ/GPTQ/INT4 repos report packed element count as parameters — 6x memory understatement, rated 'Perfect'
bugUpdated Sep 15, 2026 - #122
Detect NPUs and variant core specs?
Feature RequestHardware CompatibilityFeasibilityUpdated Sep 14, 2026 - #1036
Fit estimates don't account for Unsloth Studio's beyond-VRAM offload
enhancementHardware CompatibilityUpdated Sep 14, 2026 - #1015
[Feature]: Add a guided CLI wizard for goal-driven model recommendations
enhancementUpdated Sep 13, 2026 - #1021
`llmfit update` fetches model lists successfully but caches 0 of 237 models
bugUpdated Sep 13, 2026 - #1024
Ollama gemma4:12b installed locally but not recognized by llmfit (ollama_name mapping missing for whole Gemma 4 family)
bugUpdated Sep 13, 2026 - #1018
[Feature]: mutiple GPU = total VRAM
enhancementUpdated Sep 11, 2026 - #974
ux(tui): show estimate confidence (and prefill) in detail pane
enhancementUpdated Sep 5, 2026 - #973
fix(fit): prefer native MXFP4 quant path for gpt-oss-class MoE models
enhancementEfficiencyUpdated Aug 30, 2026 - #972
test: per-chip-class estimator accuracy CI gate (~25% median)
enhancementTestingUpdated Aug 30, 2026 - #292
For the given recommendation model, can you provide the corresponding llama.cpp runtime parameters?
questionUpdated Aug 28, 2026