All Workflows Performance Benchmarks CI Documentation GPU Tests Release

22 workflow runs

Actor

All actors osobh

Status

All status Success Failure Canceled Skipped Waiting Running Blocked Canceling

Branch

All branches main

Create status badge

rtx-backend-cuda: native index_select / index_add — HetGAT training 2.5x, inference 4.3x

#597:Commit 796b8487ad pushed by osobh

main

2026-09-12 04:09:08 +00:00

2s

View workflow file

feat(mamba): GPU-accelerated backward pass (backward_cuda)

#278:Commit 5155c081ca pushed by osobh

main

2026-08-10 14:11:31 +00:00

3s

View workflow file

feat(demos,inference): wire simulation demos to real compute; fix embedding lookup and weight-name aliases

#232:Commit e080748d88 pushed by osobh

main

2026-07-10 05:06:39 +00:00

6s

View workflow file

feat(inference): concrete EAGLE draft model + real tokenizer at the serving boundary

#228:Commit 733b02cd8b pushed by osobh

main

2026-07-10 04:54:21 +00:00

8s

View workflow file

chore(sweep): delete 43 orphaned source files; document SYCL/demo/duplication status

#221:Commit 5f32165184 pushed by osobh

main

2026-07-10 02:35:29 +00:00

4s

View workflow file

fix(deps): bump candle-core/nn/transformers 0.8->0.11 for CUDA 13.1 build

#217:Commit 522400a72b pushed by osobh

main

2026-07-10 01:10:53 +00:00

7s

View workflow file

fix(ci,rtx-tensor): resolve clippy --all-features intel-mkl conflict; gate MKL to x86_64-linux

#189:Commit e8a2036db4 pushed by osobh

main

2026-06-27 15:13:52 +00:00

6s

View workflow file

feat(batch19): Medusa heads, TIES+DARE model merging, Mixture of Depths

#173:Commit bff27c302f pushed by osobh

main

2026-06-27 15:13:19 +00:00

2h4m22s

View workflow file

feat(batch16): SOAP optimizer, lookahead decoding, SWA+SWAG

#163:Commit 54a9652041 pushed by osobh

main

2026-06-27 15:13:18 +00:00

8h28m34s

View workflow file

feat(batch15): Shampoo optimizer, beam search decoder, sliding window attention

#156:Commit 6daa20c94b pushed by osobh

main

2026-06-27 15:13:18 +00:00

8h45m40s

View workflow file

feat(batch14): Muon optimizer, logit processors, per-token activation quantization

#152:Commit e00a6da018 pushed by osobh

main

2026-06-27 15:13:17 +00:00

9h1m13s

View workflow file

Merge branch 'feat/f64-rtx-tensor'

#136:Commit 877071bb9a pushed by osobh

main

2026-06-27 15:13:17 +00:00

9h43m33s

View workflow file

feat(batch11): WSD LR scheduler, KV CPU offloading, GQA KV head expansion

#132:Commit 4ae0c34537 pushed by osobh

main

2026-06-27 15:13:17 +00:00

9h44m21s

View workflow file

feat(batch10): attention sinks (StreamingLLM), chunked prefill, per-layer LR decay

#125:Commit be3e3965b1 pushed by osobh

main

2026-06-27 15:13:16 +00:00

10h18m15s

View workflow file

feat(batch9): token merging (ToMe), grad accum per-step norm, speculative streaming

#121:Commit ef3cdb1e1a pushed by osobh

main

2026-06-27 15:13:16 +00:00

10h34m10s

View workflow file

feat(batch7): interleaved 1F1B, attention-selective checkpointing, flash decoding

#111:Commit b45a58792d pushed by osobh

main

2026-06-27 15:13:15 +00:00

11h6m8s

View workflow file

feat(batch6): windowed acceptance metrics, KV INT8 quant, col/row-parallel linear

#107:Commit 0e1d6a74b6 pushed by osobh

main

2026-06-27 15:13:15 +00:00

11h18m52s

View workflow file

feat(perf): GPU perf batch 4 — SmoothQuant INT8 forward, varlen FA, inference graph capture

#100:Commit d6769ef641 pushed by osobh

main

2026-06-27 15:13:15 +00:00

14h1m57s

View workflow file

feat(perf): GPU perf batch 2 — EAGLE-3 dynamic draft trees + GaLore-2 optimizer

#93:Commit a670045f46 pushed by osobh

main

2026-06-27 15:13:14 +00:00

14h36m28s

View workflow file

feat(perf): GPU perf batch 1 — wire GPU execution paths for FP8, FA3, CUDA Graphs, SnapKV

#89:Commit 082a50e3a0 pushed by osobh

main

2026-06-27 15:13:14 +00:00

14h36m37s

View workflow file

feat(perf): GPU perf batch 1 — FP8, FA3 Blackwell, CUDA Graphs, SnapKV, prefix cache

#85:Commit 1eb89c5b2b pushed by osobh

main

2026-06-27 15:13:13 +00:00

14h36m47s

View workflow file

style: rustfmt formatting pass on rtx-tensor and rtx-flash-attention

#81:Commit c82d26d6e7 pushed by osobh

main

2026-06-27 15:13:13 +00:00

14h37m1s

View workflow file