Skip to content

Actions: levvius/adaptive-speculative-decoding

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
28 workflow runs
28 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
Refresh docs for JointAdaSpec results and workflow CI #28: Commit 8c76b64 pushed by levvius
2m 30s main
main
2m 30s
Add JointAdaSpec Qwen 7B/1.5B experiment reports CI #27: Commit 57010bb pushed by levvius
3m 34s main
main
3m 34s
Add Qwen consensus run artifacts CI #26: Commit 0bcce61 pushed by levvius
2m 31s main
main
2m 31s
Add consensus AutoJudge ensemble drafting method CI #25: Commit bf15e33 pushed by levvius
2m 24s main
main
2m 24s
Add distributional features (hidden+dist) to AutoJudge classifier CI #24: Commit 62a590b pushed by levvius
2m 31s main
main
2m 31s
Add Llama-3 8B/3B k=8 evaluation reports (GSM8K + LiveCodeBench) CI #23: Commit 198f13d pushed by levvius
2m 30s main
main
2m 30s
Add HGBT AutoJudge backend and 48h Llama reports CI #22: Commit 40fc797 pushed by levvius
2m 28s main
main
2m 28s
Improve GitHub presentation and contributor onboarding CI #21: Commit 7a5112d pushed by levvius
2m 44s main
main
2m 44s
Calcul Model llama CI #20: Commit 9e2345e pushed by levvius
2m 44s main
main
2m 44s
Align AutoJudge C-grid presets with paper CI #19: Commit 694e397 pushed by levvius
2m 33s main
main
2m 33s
Add Meta Llama smoke run reports (2026-03-24) CI #18: Commit 4a7b46c pushed by levvius
2m 36s main
main
2m 36s
Add Llama-3 8B / Llama-3.2 3B config presets and eval wrapper CI #17: Commit 865d45f pushed by levvius
2m 36s main
main
2m 36s
Result gemma CI #16: Commit 9571a26 pushed by levvius
2m 40s main
main
2m 40s
Add Qwen2.5 7B/1.5B eval reports and Gemma-2 9B/2B config presets CI #15: Commit a1f1ce7 pushed by levvius
2m 45s main
main
2m 45s
update 06-03-2026 CI #14: Commit b719f09 pushed by levvius
2m 54s main
main
2m 54s
Align AutoJudge C grid with paper; add paper-alignment docs and file_… CI #13: Commit 8134ac0 pushed by levvius
2m 58s main
main
2m 58s
Switch paper defaults to Qwen2.5 0.5B->3B compatible pair CI #12: Commit 700c04f pushed by levvius
3m 5s main
main
3m 5s
docs(deps): refresh README/CODEX and pin autojudge dependency CI #11: Commit 2f1400a pushed by levvius
3m 36s main
main
3m 36s
Fix RTX50 Docker stack and CUDA arch compatibility checks CI #10: Commit aa2cfa2 pushed by levvius
3m 19s main
main
3m 19s
Add CUDA preflight and Docker GPU diagnostics targets CI #9: Commit ee068cc pushed by levvius
1m 40s main
main
1m 40s
Add Docker build fallback for BuildKit snapshot failures CI #8: Commit 1942c58 pushed by levvius
1m 53s main
main
1m 53s
Avoid passing null quantization config to HF from_pretrained CI #7: Commit ad09749 pushed by levvius
1m 43s main
main
1m 43s
Ignore quant overrides for native-quantized GPT-OSS models CI #6: Commit 1a77d23 pushed by levvius
1m 39s main
main
1m 39s
Handle GPT-OSS quantization config conflicts in HF loader CI #5: Commit 5d48719 pushed by levvius
1m 46s main
main
1m 46s
Fix Dockerfile.gpu when CUDA base image lacks pip CI #4: Commit 0e290c8 pushed by levvius
1m 40s main
main
1m 40s