Skip to content

Actions: huggingface/candle

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
2,500+ workflow runs
2,500+ 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
fix(gguf): bound tensor size against file before allocating Continuous integration #8375: Pull request #3588 synchronize by pjdurden
Action required pjdurden:fix/gguf-tensorinfo-alloc-overflow
pjdurden:fix/gguf-tensorinfo-alloc-overflow
Action required
fix(gguf): bound tensor size against file before allocating CI / cuda #5291: Pull request #3588 synchronize by pjdurden
Action required pjdurden:fix/gguf-tensorinfo-alloc-overflow
pjdurden:fix/gguf-tensorinfo-alloc-overflow
Action required
Fix causal attention in candle-flash-attn-v3: NULL tile_count_semaphore crash (dense) + silently non-causal varlen CI / cuda #5290: Pull request #3606 opened by Xueying-VirtueAI
Action required Xueying-VirtueAI:fa3-fix-causal
Xueying-VirtueAI:fa3-fix-causal
Action required
Fix causal attention in candle-flash-attn-v3: NULL tile_count_semaphore crash (dense) + silently non-causal varlen Continuous integration #8374: Pull request #3606 opened by Xueying-VirtueAI
Action required Xueying-VirtueAI:fa3-fix-causal
Xueying-VirtueAI:fa3-fix-causal
Action required
fix(transformers): adds support for llamas head_dim config Continuous integration #8373: Pull request #3602 opened by suniastar
Action required suniastar:fix/llama-non-standard-head-dim
suniastar:fix/llama-non-standard-head-dim
Action required
fix(transformers): adds support for llamas head_dim config CI / cuda #5289: Pull request #3602 opened by suniastar
Action required suniastar:fix/llama-non-standard-head-dim
suniastar:fix/llama-non-standard-head-dim
Action required
fix(kernels): Compilation errors on Turing Cards (Compute Cap 7.5) Continuous integration #8372: Pull request #3601 opened by suniastar
Action required suniastar:fix/cuda-on-turing
suniastar:fix/cuda-on-turing
Action required
fix(kernels): Compilation errors on Turing Cards (Compute Cap 7.5) CI / cuda #5288: Pull request #3601 opened by suniastar
Action required suniastar:fix/cuda-on-turing
suniastar:fix/cuda-on-turing
Action required
fix(transformers): adds support for non-default head_dim in llama Continuous integration #8370: Pull request #3593 synchronize by suniastar
Action required suniastar:feat/llama-non-standard-head-dim
suniastar:feat/llama-non-standard-head-dim
Action required
fix(transformers): adds support for non-default head_dim in llama CI / cuda #5286: Pull request #3593 synchronize by suniastar
Action required suniastar:feat/llama-non-standard-head-dim
suniastar:feat/llama-non-standard-head-dim
Action required
build(cuda): register compute capability cfgs and optional legacy BF16/FP8 emulation Continuous integration #8369: Pull request #2704 synchronize by haricot
Action required haricot:bf16_candle
haricot:bf16_candle
Action required
build(cuda): register compute capability cfgs and optional legacy BF16/FP8 emulation CI / cuda #5285: Pull request #2704 synchronize by haricot
Action required haricot:bf16_candle
haricot:bf16_candle
Action required
Optimize CUDA kernels CI / cuda #5284: Pull request #3600 opened by guoqingbao
Action required guoqingbao:cuda_kernels
guoqingbao:cuda_kernels
Action required
Optimize CUDA kernels Continuous integration #8368: Pull request #3600 opened by guoqingbao
Action required guoqingbao:cuda_kernels
guoqingbao:cuda_kernels
Action required
Metal: fix SDPA storage offsets Continuous integration #8367: Pull request #3599 opened by EricLBuehler
12m 52s metal-sdpa-byte-offsets
metal-sdpa-byte-offsets
12m 52s
Metal: fix SDPA storage offsets CI / cuda #5283: Pull request #3599 opened by EricLBuehler
9m 23s metal-sdpa-byte-offsets
metal-sdpa-byte-offsets
9m 23s
Metal SDPA: pass storage offsets in bytes, not elements Secret Leaks #1505: Commit d2afd7f pushed by EricLBuehler
23s metal-sdpa-byte-offsets
metal-sdpa-byte-offsets
23s
Parameter cache for CUDA kernel launch Continuous integration #8366: Pull request #3598 opened by guoqingbao
Action required guoqingbao:parameter_cache
guoqingbao:parameter_cache
Action required
Parameter cache for CUDA kernel launch CI / cuda #5282: Pull request #3598 opened by guoqingbao
Action required guoqingbao:parameter_cache
guoqingbao:parameter_cache
Action required
Fix 128x softmax_lse over-allocation in the dense flash-attn-v3 path CI / cuda #5281: Pull request #3597 opened by Xueying-VirtueAI
Action required Xueying-VirtueAI:fa3-fix-lse-overalloc
Xueying-VirtueAI:fa3-fix-lse-overalloc
Action required
Fix 128x softmax_lse over-allocation in the dense flash-attn-v3 path Continuous integration #8365: Pull request #3597 opened by Xueying-VirtueAI
Action required Xueying-VirtueAI:fa3-fix-lse-overalloc
Xueying-VirtueAI:fa3-fix-lse-overalloc
Action required
candle-core: add Device::enable_peer_access for cross-CUDA P2P transfers CI / cuda #5280: Pull request #3525 synchronize by toddwbucy
Action required toddwbucy:feat/peer-access-upstream
toddwbucy:feat/peer-access-upstream
Action required
candle-core: add Device::enable_peer_access for cross-CUDA P2P transfers Continuous integration #8364: Pull request #3525 synchronize by toddwbucy
-1s toddwbucy:feat/peer-access-upstream
toddwbucy:feat/peer-access-upstream
-1s
flash-attn: launch kernels on the caller's CUDA stream CI / cuda #5279: Pull request #3596 opened by jnises
Action required jnises:flash-attn-caller-stream
jnises:flash-attn-caller-stream
Action required
flash-attn: launch kernels on the caller's CUDA stream Continuous integration #8363: Pull request #3596 opened by jnises
Action required jnises:flash-attn-caller-stream
jnises:flash-attn-caller-stream
Action required