Skip to content

Actions: PrismML-Eng/llama.cpp

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
2,500+ workflow runs
2,500+ workflow runs

Filter by Workflow

⌥ + click/return to exclude

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
cuda: add PTQ1_0 support Update Operations Documentation #39: Commit a294e8d pushed by bri-prism
34s ptq1_0-cuda
ptq1_0-cuda
34s
cuda: add PTQ1_0 support Python check requirements.txt #37: Commit a294e8d pushed by bri-prism
2m 52s ptq1_0-cuda
ptq1_0-cuda
2m 52s
cuda: add PTQ1_0 support Python Type-Check #218: Commit a294e8d pushed by bri-prism
3m 16s ptq1_0-cuda
ptq1_0-cuda
3m 16s
cuda: add PTQ1_0 support Check Pre-Tokenizer Hashes #63: Commit a294e8d pushed by bri-prism
1m 4s ptq1_0-cuda
ptq1_0-cuda
1m 4s
ggml: add PTQ1_0, ternary at group 128 Update Operations Documentation #38: Commit 7ce416f pushed by bri-prism
30s ptq1_0-base
ptq1_0-base
30s
ggml: add PTQ1_0, ternary at group 128 Check Pre-Tokenizer Hashes #62: Commit 7ce416f pushed by bri-prism
1m 2s ptq1_0-base
ptq1_0-base
1m 2s
ggml: add PTQ1_0, ternary at group 128 Python check requirements.txt #36: Commit 7ce416f pushed by bri-prism
2m 52s ptq1_0-base
ptq1_0-base
2m 52s
ggml: add PTQ1_0, ternary at group 128 Python Type-Check #217: Commit 7ce416f pushed by bri-prism
3m 26s ptq1_0-base
ptq1_0-base
3m 26s
cuda: prefetch the next PTQ1_0 mat-vec's weights into L2 from the last CTAs Server #429: Pull request #314 opened by sb32445
Action required sb32445:pr/ptq1-l2-prefetch
sb32445:pr/ptq1-l2-prefetch
Action required
cuda: prefetch the next PTQ1_0 mat-vec's weights into L2 from the last CTAs Pull Request Labeler #459: Pull request #314 opened by sb32445
22s
22s
common : take the top-k straight from the logits when top-k starts the chain Server #428: Pull request #313 opened by sb32445
Action required sb32445:pr/sampler-topk-from-logits
sb32445:pr/sampler-topk-from-logits
Action required
common : take the top-k straight from the logits when top-k starts the chain Pull Request Labeler #458: Pull request #313 opened by sb32445
17s
17s
server : reuse the buffers of evicted prompt checkpoints Server #427: Pull request #312 opened by sb32445
Action required sb32445:pr/server-ckpt-buffer-reuse
sb32445:pr/server-ckpt-buffer-reuse
Action required
server : reuse the buffers of evicted prompt checkpoints Pull Request Labeler #457: Pull request #312 opened by sb32445
21s
21s
kv-cache : limit seq_rm to the used cell range Server #426: Pull request #311 opened by sb32445
Action required sb32445:pr/kv-seq-rm-bound
sb32445:pr/kv-seq-rm-bound
Action required
kv-cache : limit seq_rm to the used cell range Pull Request Labeler #456: Pull request #311 opened by sb32445
16s
16s
cuda: gate (SwiGLU) fused PTQ1_0 mat-vec for 2-4 columns Server #425: Pull request #310 opened by sb32445
Action required sb32445:pr/ptq1-gate-up-fuse-mc
sb32445:pr/ptq1-gate-up-fuse-mc
Action required
cuda: gate (SwiGLU) fused PTQ1_0 mat-vec for 2-4 columns Pull Request Labeler #455: Pull request #310 opened by sb32445
16s
16s
cuda: use the transposing concat kernel on all GPUs, not only on GB10 Server #424: Pull request #309 opened by sb32445
Action required sb32445:pr/concat-transpose-all-gpus
sb32445:pr/concat-transpose-all-gpus
Action required
cuda: use the transposing concat kernel on all GPUs, not only on GB10 Pull Request Labeler #454: Pull request #309 opened by sb32445
16s
16s
cuda: smaller KV tile for the 8-column MMA flash attention config (head size 256) Pull Request Labeler #453: Pull request #308 opened by sb32445
18s
18s
cuda: use the MMA flash attention kernel for GQA above 4 with quantized K/V on Ada Server #423: Pull request #307 opened by sb32445
Action required sb32445:pr/fattn-gqa-mma
sb32445:pr/fattn-gqa-mma
Action required
cuda: use the MMA flash attention kernel for GQA above 4 with quantized K/V on Ada Pull Request Labeler #452: Pull request #307 opened by sb32445
15s
15s
cuda: PQ2_0 mat-vec kernel for 3-8 columns on Ada Server #422: Pull request #306 opened by sb32445
Action required sb32445:pr/pq2_0-multicol
sb32445:pr/pq2_0-multicol
Action required
cuda: PQ2_0 mat-vec kernel for 3-8 columns on Ada Pull Request Labeler #451: Pull request #306 opened by sb32445
17s
17s