Skip to content

Actions: protobird-git/ai-edge-torch

Actions

Run Post Merge

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
run_post_merge.yml
33 workflow runs
33 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Bump ai-edge-torch nightly version Run Post Merge #33: Commit 20b49d2 pushed by protobird-git
1d 0h 0m 7s main
main
1d 0h 0m 7s
Remove kv_cache_max_len from ModelConfig. Run Post Merge #32: Commit 93edc84 pushed by protobird-git
1d 0h 0m 9s main
main
1d 0h 0m 9s
fix quantization_config error when running convert with new quantizat… Run Post Merge #31: Commit daa06ff pushed by protobird-git
1d 0h 0m 8s main
main
1d 0h 0m 8s
Call sdpa_with_kv_update from non-experimental attention Run Post Merge #30: Commit f46489c pushed by protobird-git
1d 0h 0m 10s main
main
1d 0h 0m 10s
Support torch.ops.aten.sym_size.int and torch Tensor.reshape. Run Post Merge #29: Commit 9d83b50 pushed by protobird-git
1d 0h 0m 10s main
main
1d 0h 0m 10s
Reduce number of retracing (Canonicalization) in conversion fx pass p… Run Post Merge #28: Commit 0db3444 pushed by protobird-git
1d 0h 0m 10s main
main
1d 0h 0m 10s
Use explicit fake impl for torch.ops.tfl.reshape Run Post Merge #27: Commit 9625f16 pushed by protobird-git
1d 0h 0m 10s main
main
1d 0h 0m 10s
Support Phi-4 model Run Post Merge #26: Commit 29311c6 pushed by protobird-git
1d 15h 39m 1s main
main
1d 15h 39m 1s
Internal code change. Run Post Merge #25: Commit 883075e pushed by protobird-git
1d 0h 22m 5s main
main
1d 0h 22m 5s
Set enable_hlfb to true for PaliGemma image encoder Run Post Merge #24: Commit f242ee3 pushed by protobird-git
1d 15h 57m 18s main
main
1d 15h 57m 18s
Full stack of Qwen2.5-VL model. Run Post Merge #23: Commit 48d225a pushed by protobird-git
1d 2h 9m 16s main
main
1d 2h 9m 16s
Fix broken PaliGemma2 Run Post Merge #22: Commit 45f18d2 pushed by protobird-git
1d 21h 4m 44s main
main
1d 21h 4m 44s
Pass mask and export config correctly in PaliGemma's decoder. Run Post Merge #21: Commit 21d2732 pushed by protobird-git
1d 0h 55m 25s main
main
1d 0h 55m 25s
Add decoder of Qwen2.5-VL model. Run Post Merge #20: Commit 78bbac2 pushed by protobird-git
1d 16h 15m 40s main
main
1d 16h 15m 40s
GPU model prefill sequence length minimum is 32 Run Post Merge #19: Commit f87e4fa pushed by protobird-git
1d 2h 8m 28s main
main
1d 2h 8m 28s
update requirements.txt with torch 2.6.0 Run Post Merge #18: Commit 7f2b452 pushed by protobird-git
1d 16h 12m 13s main
main
1d 16h 12m 13s
Fix RemoveSDPACompositeZeroMaskPass with new decomp rules Run Post Merge #17: Commit 4844c1c pushed by protobird-git
1d 20h 10m 25s main
main
1d 20h 10m 25s
Add DeepSeek-R1-Distill-Qwen as an example Run Post Merge #16: Commit 42eb2ef pushed by protobird-git
1d 18h 39m 48s main
main
1d 18h 39m 48s
internal changes only Run Post Merge #15: Commit 243fae8 pushed by protobird-git
1d 14h 32m 52s main
main
1d 14h 32m 52s
torch 2.6.0 upgrade patch Run Post Merge #14: Commit 6b0713b pushed by protobird-git
1d 15h 36m 23s main
main
1d 15h 36m 23s
Enable PaliGemma conversion tests Run Post Merge #13: Commit 34296a8 pushed by protobird-git
1d 20h 1m 11s main
main
1d 20h 1m 11s
Handle mask properly in gemma2 Run Post Merge #12: Commit c741a95 pushed by protobird-git
1d 22h 34m 42s main
main
1d 22h 34m 42s
Fix broken Phi2 and PaliGemma1 Run Post Merge #11: Commit c8d0b19 pushed by protobird-git
1d 13h 45m 35s main
main
1d 13h 45m 35s
fix phi-3 conversion ckpt path. Run Post Merge #10: Commit 58f1c71 pushed by protobird-git
1d 0h 4m 28s main
main
1d 0h 4m 28s
Convert PaliGemma2 Run Post Merge #9: Commit 562c93d pushed by protobird-git
1d 19h 17m 4s main
main
1d 19h 17m 4s