Skip to content

Actions: AmesianX/TurboQuant

Actions

AI review (issues)

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
ai-issues.yml
9 workflow runs
9 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
win build? AI review (issues) #9: Issue #22 opened by jarkevithwlad
1d 0h 0m 2s
1d 0h 0m 2s
plz make it support mtp too AI review (issues) #8: Issue #21 opened by xueqing0622
1d 0h 0m 2s
1d 0h 0m 2s
Misc. bug: ggml\src\ggml-cuda\fattn.cu:744: fatal error AI review (issues) #7: Issue #20 opened by SPn00b
1d 0h 0m 2s
1d 0h 0m 2s
Reproducibility: tbq4/tbqp pipeline is non-deterministic even at fixed --seed (0.4% embedding drift between identical runs) AI review (issues) #6: Issue #19 opened by cutlerbenjamin1-cmd
1d 0h 0m 2s
1d 0h 0m 2s
Eval bug: llama-server crashes without error if using any turboquant cache type. works normally with q4_0 AI review (issues) #5: Issue #18 opened by infinity7117
1d 0h 0m 2s
1d 0h 0m 2s
Eval bug: gemma-4-26b-a4b-it-turbo using tbq4_0 is insane (and gemma3 fails to load correctly) AI review (issues) #4: Issue #17 opened by zekrom-vale
2h 20m 31s
2h 20m 31s
Eval bug: Using --cache-type-v with any tbq causes the model to break AI review (issues) #3: Issue #16 opened by StreamVersus
2h 44m 35s
2h 44m 35s
Eval bug: gemma output <unused24><unused24><unused24><unused24> AI review (issues) #2: Issue #15 opened by jarkevithwlad
1d 0h 0m 3s
1d 0h 0m 3s
Compile bug: warning: no usable GPU found Build When DGGML_HIP=ON AI review (issues) #1: Issue #14 opened by baramofme
1d 0h 0m 1s
1d 0h 0m 1s