Skip to content

Actions: dphnAI/aphrodite-engine

Actions

Deploy to GitHub Pages

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
deploy.yml
368 workflow runs
368 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
feat: add dflash support for gemma-4 (#1673) Deploy to GitHub Pages #984: Commit 14e8de1 pushed by AlpinDale
59s main
main
59s
fix: gemma-4 with exllamav3 (#1672) Deploy to GitHub Pages #983: Commit c16f370 pushed by AlpinDale
52s main
main
52s
fix: logging messages in metal (#1671) Deploy to GitHub Pages #982: Commit acd0661 pushed by AlpinDale
47s main
main
47s
perf: optimize GDN performance on Metal (#1670) Deploy to GitHub Pages #981: Commit f980722 pushed by AlpinDale
55s main
main
55s
chore: optimize metal backend performance (#1669) Deploy to GitHub Pages #980: Commit 643efa6 pushed by AlpinDale
56s main
main
56s
feat: add native Metal support (#1668) Deploy to GitHub Pages #979: Commit c19f1fe pushed by AlpinDale
47s main
main
47s
chore: sync to upstream 985961345a13f3e3bb15d29c94b011ba9a6b858b (#1666) Deploy to GitHub Pages #978: Commit 18f852d pushed by AlpinDale
50s main
main
50s
feat: add silu clamp limit to shared expert for DeepSeek-V4 (#1665) Deploy to GitHub Pages #977: Commit e958628 pushed by AlpinDale
42s main
main
42s
fix: remove invalid deepstack boundary check for Qwen3-VL (#1664) Deploy to GitHub Pages #976: Commit 6eeee1b pushed by AlpinDale
57s main
main
57s
chore: share dequant buffers in TurboQuant to save memory (#1661) Deploy to GitHub Pages #975: Commit b42eeec pushed by AlpinDale
49s main
main
49s
feat: FP8 ViT Attention w/ FlashInfer (#1660) Deploy to GitHub Pages #974: Commit 1241957 pushed by AlpinDale
51s main
main
51s
fix: cap SWA/chunked-local runtime admission to startup pool-sizing b… Deploy to GitHub Pages #973: Commit 42769da pushed by AlpinDale
47s main
main
47s
fix: rejection sampling acceptance rate in MRv2 (#1658) Deploy to GitHub Pages #972: Commit f48068d pushed by AlpinDale
44s main
main
44s
fix: auto-disable expandable_segments around cumem memory pool (#1657) Deploy to GitHub Pages #971: Commit 939683e pushed by AlpinDale
50s main
main
50s
perf: exl3 optims with guarded MoE down tuning (#1656) Deploy to GitHub Pages #970: Commit 6fe6276 pushed by AlpinDale
53s main
main
53s
perf: exl3 decode kernel optimization experiments (#1655) Deploy to GitHub Pages #969: Commit 6c59bc7 pushed by AlpinDale
44s main
main
44s
fix: normalize nested args in DeepSeek DSML (#1654) Deploy to GitHub Pages #968: Commit c005aac pushed by AlpinDale
55s main
main
55s
fix: TRT-LLM MXFP4 MoE compile for DeepSeek-V4 (#1653) Deploy to GitHub Pages #967: Commit 74714e2 pushed by AlpinDale
52s main
main
52s
perf: EXL3 performance tuning on GeForce Blackwell (#1652) Deploy to GitHub Pages #966: Commit 3917cd8 pushed by AlpinDale
51s main
main
51s
feat: implement DeepSeek-V4 model (#1651) Deploy to GitHub Pages #965: Commit 61aad7c pushed by AlpinDale
51s main
main
51s
fix: disable sequence parallelism for piecewise compilation (#1650) Deploy to GitHub Pages #964: Commit 19b1e78 pushed by AlpinDale
54s main
main
54s
fix: offload all KV blocks when doing prefill in P/D (#1649) Deploy to GitHub Pages #963: Commit f2764f3 pushed by AlpinDale
53s main
main
53s
fix: size the MNNVL workspace for flashinfer to EP group (#1648) Deploy to GitHub Pages #962: Commit 1d9d9b5 pushed by AlpinDale
55s main
main
55s
refactor: MoE LoRA Deploy to GitHub Pages #961: Commit 87fd59b pushed by AlpinDale
50s main
main
50s
fix: DSML token leakage in DeepSeek-V4 and 3.2 (#1647) Deploy to GitHub Pages #960: Commit 8556a53 pushed by AlpinDale
45s main
main
45s