Skip to content

Actions: ORippler/llama.cpp

Actions

Python check requirements.txt

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
python-check-requirements.yml
34 workflow runs
34 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
LC to overlap with following kernels Python check requirements.txt #34: Commit b508845 pushed by ORippler
1d 0h 0m 2s osimons/enroll_mul_mat_vec_q_moe_into_PDL
osimons/enroll_mul_mat_vec_q_moe_into_PDL
1d 0h 0m 2s
CUDA: restrict PDL to CTK >= 12.3 due to MSVC issues Python check requirements.txt #33: Commit fa5673c pushed by ORippler
1d 0h 0m 2s osimons/pdl_version_fix
osimons/pdl_version_fix
1d 0h 0m 2s
CUDA: Check PTX version on host side to guard PDL dispatch Python check requirements.txt #32: Commit bcde351 pushed by ORippler
6m 18s osimons/ptx_pdl_checks_cuda
osimons/ptx_pdl_checks_cuda
6m 18s
Move check inside scales branch Python check requirements.txt #31: Commit f3feefa pushed by ORippler
7m 32s osimons/cuda_use_in_scale
osimons/cuda_use_in_scale
7m 32s
CUDA: Continue directly including cuda/iterator Python check requirements.txt #30: Commit 4285665 pushed by ORippler
6m 17s osimons/cuda_iterator_cont
osimons/cuda_iterator_cont
6m 17s
CUDA: directly include cuda/iterator Python check requirements.txt #29: Commit 8d0cd8b pushed by ORippler
6m 25s osimons/cuda_explicit_iterator_import
osimons/cuda_explicit_iterator_import
6m 25s
WIP fastdiv_u64 implementation in CUDA Python check requirements.txt #28: Commit c93a762 pushed by ORippler
5m 31s osimons/fdiv_u64_poc
osimons/fdiv_u64_poc
5m 31s
CUDA: better coalesce data-access for contiguous concat Python check requirements.txt #27: Commit a027d4f pushed by ORippler
5m 27s osimons/CUDA_faster_cont_concat
osimons/CUDA_faster_cont_concat
5m 27s
Initial WIP commit Python check requirements.txt #26: Commit cd243c5 pushed by ORippler
5m 44s osimons/derived_tensor_poc
osimons/derived_tensor_poc
5m 44s
Optimize data-movement of load_tiles_nvfp4_producer Python check requirements.txt #25: Commit fb42c25 pushed by ORippler
6m 21s osimons/warp-specialization-mmq-wip
osimons/warp-specialization-mmq-wip
6m 21s
CUDA: Limit DeviceSegmentedSort to immediate mode Python check requirements.txt #24: Commit 51a86d5 pushed by ORippler
6m 34s osimons/cuda_fix_argsort_graph_capture
osimons/cuda_fix_argsort_graph_capture
6m 34s
CUDA: Fix CUB's argsort when nrows % block_size == 0 CCCL < 3.1 Python check requirements.txt #23: Commit 28c3c54 pushed by ORippler
4m 59s osimons/cuda_fix_argsort_ceildiv_for_cccl_lt_3dot1
osimons/cuda_fix_argsort_ceildiv_for_cccl_lt_3dot1
4m 59s
WIP showing that data-access is the limitation in PP Python check requirements.txt #22: Commit d7896a7 pushed by ORippler
6m 32s osimons/gdx-pp-optimizations
osimons/gdx-pp-optimizations
6m 32s
Apply code-formating Python check requirements.txt #21: Commit e1cfcf8 pushed by ORippler
5m 5s osimons/gated_delta_net_improvements
osimons/gated_delta_net_improvements
5m 5s
CUBLAS: Use FP32 accumulation also for VOLTA TCs Python check requirements.txt #20: Commit 4eee426 pushed by ORippler
5m 26s osimons/add_bf16_cublas_path
osimons/add_bf16_cublas_path
5m 26s
Add WIP implementation on cuda graph logic Python check requirements.txt #19: Commit f39a785 pushed by ORippler
6m 36s osimons/update_cuda_graph_heuristics
osimons/update_cuda_graph_heuristics
6m 36s
Apply code-formater Python check requirements.txt #18: Commit 28177e3 pushed by ORippler
5m 43s osimons/cuda_graphs_qwen3_next
osimons/cuda_graphs_qwen3_next
5m 43s
Do not mutate cgraph for fused ADDs Python check requirements.txt #17: Commit 79f5eaf pushed by ORippler
5m 36s osimons/fix_multi_add
osimons/fix_multi_add
5m 36s
Bump consec_update threshold by 1 to enable CG on qwen3-next Python check requirements.txt #16: Commit 2a59644 pushed by ORippler
5m 31s osimons/cuda_graphs_qwen3_next
osimons/cuda_graphs_qwen3_next
5m 31s
CUDA: Update CCCL-tag for 3.2 to final release from RC Python check requirements.txt #15: Commit d8da435 pushed by ORippler
5m 40s master
master
5m 40s
Fix rope_norm Python check requirements.txt #14: Commit 3fa5110 pushed by ORippler
7m 6s osimons/cuda_fix_non-contig_rope
osimons/cuda_fix_non-contig_rope
7m 6s
CUDA: Fix loop unrolling for BW in mul_mat_q_stream_k_fixup Python check requirements.txt #13: Commit b056b05 pushed by ORippler
3m 13s osimons/fix_bw_mmq_fixup_kernel
osimons/fix_bw_mmq_fixup_kernel
3m 13s
CUDA: Fix builds for older CCCL versions by ifdefing strided_iterator Python check requirements.txt #12: Commit f8b1071 pushed by ORippler
3m 32s osimons/ifdef_strided_iterator
osimons/ifdef_strided_iterator
3m 32s
Remove unnecessary include from top-k.cu Python check requirements.txt #11: Commit 90c162a pushed by ORippler
3m 30s osimons/cuda_iterator_in_argsort
osimons/cuda_iterator_in_argsort
3m 30s
Use two_stage_warp_reduce also in softmax kernel, move smem out of it Python check requirements.txt #10: Commit 1ebe58d pushed by ORippler
3m 24s osimons/factor_out_two_stage_warp_reductions
osimons/factor_out_two_stage_warp_reductions
3m 24s