Skip to content

ggml: implement quantized KV cache for FA #1508

ggml: implement quantized KV cache for FA

ggml: implement quantized KV cache for FA #1508

Triggered via pull request May 18, 2024 20:34
Status Success
Total duration 14h 27m 12s
Artifacts 3

bench.yml

on: pull_request_target
Matrix: bench-server-baseline
Fit to window
Zoom out
Zoom in

Annotations

15 warnings
bench-server-baseline (phi-2, f16)
Node.js 16 actions are deprecated. Please update the following actions to use Node.js 20: Sibz/github-status-action@v1, devicons/public-upload-to-imgur@v2.2.2. For more information see: https://github.blog/changelog/2023-09-22-github-actions-transitioning-from-node-16-to-node-20/.
bench-server-baseline (phi-2, f16)
The following actions uses node12 which is deprecated and will be forced to run on node16: Sibz/github-status-action@v1, devicons/public-upload-to-imgur@v2.2.2. For more info: https://github.blog/changelog/2023-06-13-github-actions-all-actions-will-run-on-node16-instead-of-node12-by-default/
bench-server-baseline (phi-2, f16)
Restore cache failed: Dependencies file is not found in /github-runner/_work/llama.cpp/llama.cpp. Supported file pattern: go.sum
bench-server-baseline (phi-2, f16)
The `set-output` command is deprecated and will be disabled soon. Please upgrade to using Environment Files. For more information see: https://github.blog/changelog/2022-10-11-github-actions-deprecating-save-state-and-set-output-commands/
bench-server-baseline (phi-2, f16)
The `set-output` command is deprecated and will be disabled soon. Please upgrade to using Environment Files. For more information see: https://github.blog/changelog/2022-10-11-github-actions-deprecating-save-state-and-set-output-commands/
bench-server-baseline (phi-2, q4_0)
Node.js 16 actions are deprecated. Please update the following actions to use Node.js 20: Sibz/github-status-action@v1, devicons/public-upload-to-imgur@v2.2.2. For more information see: https://github.blog/changelog/2023-09-22-github-actions-transitioning-from-node-16-to-node-20/.
bench-server-baseline (phi-2, q4_0)
The following actions uses node12 which is deprecated and will be forced to run on node16: Sibz/github-status-action@v1, devicons/public-upload-to-imgur@v2.2.2. For more info: https://github.blog/changelog/2023-06-13-github-actions-all-actions-will-run-on-node16-instead-of-node12-by-default/
bench-server-baseline (phi-2, q4_0)
Restore cache failed: Dependencies file is not found in /github-runner/_work/llama.cpp/llama.cpp. Supported file pattern: go.sum
bench-server-baseline (phi-2, q4_0)
The `set-output` command is deprecated and will be disabled soon. Please upgrade to using Environment Files. For more information see: https://github.blog/changelog/2022-10-11-github-actions-deprecating-save-state-and-set-output-commands/
bench-server-baseline (phi-2, q4_0)
The `set-output` command is deprecated and will be disabled soon. Please upgrade to using Environment Files. For more information see: https://github.blog/changelog/2022-10-11-github-actions-deprecating-save-state-and-set-output-commands/
bench-server-baseline (phi-2, q8_0)
Node.js 16 actions are deprecated. Please update the following actions to use Node.js 20: Sibz/github-status-action@v1, devicons/public-upload-to-imgur@v2.2.2. For more information see: https://github.blog/changelog/2023-09-22-github-actions-transitioning-from-node-16-to-node-20/.
bench-server-baseline (phi-2, q8_0)
The following actions uses node12 which is deprecated and will be forced to run on node16: Sibz/github-status-action@v1, devicons/public-upload-to-imgur@v2.2.2. For more info: https://github.blog/changelog/2023-06-13-github-actions-all-actions-will-run-on-node16-instead-of-node12-by-default/
bench-server-baseline (phi-2, q8_0)
The `set-output` command is deprecated and will be disabled soon. Please upgrade to using Environment Files. For more information see: https://github.blog/changelog/2022-10-11-github-actions-deprecating-save-state-and-set-output-commands/
bench-server-baseline (phi-2, q8_0)
The `set-output` command is deprecated and will be disabled soon. Please upgrade to using Environment Files. For more information see: https://github.blog/changelog/2022-10-11-github-actions-deprecating-save-state-and-set-output-commands/
bench-server-baseline (phi-2, q8_0)
Restore cache failed: Dependencies file is not found in /github-runner/_work/llama.cpp/llama.cpp. Supported file pattern: go.sum

Artifacts

Produced during runtime
Name Size
bench-server-bench-server-baseline-Standard_NC4as_T4_v3-phi-2-f16
195 KB
bench-server-bench-server-baseline-Standard_NC4as_T4_v3-phi-2-q4_0
198 KB
bench-server-bench-server-baseline-Standard_NC4as_T4_v3-phi-2-q8_0
194 KB