Skip to content

Pull requests: huggingface/candle

New pull request New
139 Open 2,075 Closed
139 Open 2,075 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[Metal] Make fast math mode optional
#3205 opened Nov 20, 2025 by ivarflakstad Loading…
1
4
Update pyo3
#3202 opened Nov 19, 2025 by ivarflakstad Loading…
Metal: bound temporary buffer cache and prevent runaway memory usage on large softmax/broadcast/matmul workloads
#3197 opened Nov 17, 2025 by TimmyOVO Loading…
26
Add more misc. changes from candle fork
#3196 opened Nov 17, 2025 by EricLBuehler Loading…
24
Add dummy dtypes
#3195 opened Nov 17, 2025 by EricLBuehler Loading…
3 tasks done
52
Add Provence model implementation
#3191 opened Nov 17, 2025 by matthewhaynesonline • Draft
Generic KvCache
#3188 opened Nov 14, 2025 by ivarflakstad Loading…
15
Add SmolLM3: Full and Quantized Implementation
#3180 opened Nov 12, 2025 by DrJesseGlass Loading…
7 tasks done
Draft: QuantizedDType for normal tensors
#3174 opened Nov 9, 2025 by LLukas22 • Draft
Use squeeze in tensor.get()
#3163 opened Nov 5, 2025 by janimo Loading…
1
Add nll_with_ignore and cross_entropy_with_ignore
#3161 opened Nov 3, 2025 by embedding-shapes Loading…
example: add quantized qwen3 wasm with SIMD optimizations
#3159 opened Oct 31, 2025 by DrJesseGlass Loading…
add padding support for max pool
#3144 opened Oct 22, 2025 by donjuanplatinum Loading…
6
feat: manual release workflow
#3128 opened Oct 11, 2025 by drbh Loading…
3
[Metal] Improved get_strided_index
#3113 opened Oct 4, 2025 by ivarflakstad • Draft
Neon CpuF16 and CpuBF16 implementations
#3110 opened Oct 3, 2025 by ivarflakstad • Draft
Updated candle-book
#3098 opened Sep 29, 2025 by little-world Loading…
3
[WIP] Implement Tensor::unfold(dim, size, step): 1-dim unfolded window views.
#3091 opened Sep 22, 2025 by crutcher Loading…
5
eliminate build dependency on installed protoc, re-export prost::Message
#3086 opened Sep 15, 2025 by dvdsk Loading…
2
[Draft] working on moving global scalar ops out of storage
#3084 opened Sep 12, 2025 by greenrazer • Draft
[WIP] CUDA graph support
#3083 opened Sep 10, 2025 by ahadnagy • Draft
Add 2 Onnx operators and constant mode to Pad
#3082 opened Sep 10, 2025 by dvdsk Loading…
Snac support
#3075 opened Sep 6, 2025 by maximizemaxwell Loading…
2
[Draft] Enable CPU multithreading in WASM with Rayon
#3063 opened Aug 23, 2025 by lucky-bai Loading…
Rev safetensors to 0.6, float8 to 0.3
#3049 opened Aug 9, 2025 by zackangelo Loading…
1
ProTip! Mix and match filters to narrow down what you’re looking for.