summaryrefslogtreecommitdiff
path: root/candle-core/benches
Commit message (Expand)AuthorAgeFilesLines
* Add support for Llama 3.1 (#2359)Eric Buehler2024-07-264-7/+7
* Metal Unary: Add benchmarks and process kernels in a tile based fashion (#2056)Thomas Santerre2024-04-213-0/+51
* Add benchmarks for qmatmul operations (#2048)Thomas Santerre2024-04-133-0/+74
* Add support for conv_transpose2d on Metal backend (#1903)Thomas Santerre2024-03-213-1/+62
* Merge branch 'main' into ivarflakstad/metal-prngIvar Flakstad2024-01-144-1/+115
|\
| * Fix format. (#1576)Nicolas Patry2024-01-121-1/+5
| * Metal: Activate bfloat affine and add benchmark (#1543)ivarflakstad2024-01-123-1/+45
| * Metal: f16 and bf16 where_cond + benchmark (#1545)ivarflakstad2024-01-123-1/+66
* | Merge branch 'main' into ivarflakstad/metal-prngIvar Flakstad2024-01-123-38/+57
|\|
| * Seperate benchmarks by enabled features (#1538)ivarflakstad2024-01-113-12/+81
* | Updated feature separated benchmarksIvar Flakstad2024-01-093-21/+14
* | Merge branch 'ivarflakstad/seperate-benchmarks-by-feature' into ivarflakstad/...Ivar Flakstad2024-01-093-10/+65
|\ \
| * | Improve benchmarks layoutIvar Flakstad2024-01-093-5/+8
| * | Avoid some unnecessary returns.Laurent2024-01-081-4/+4
| * | Remove allow pragmaIvar Flakstad2024-01-082-6/+2
| * | Use cfg to seperate benchmark results based on featuresIvar Flakstad2024-01-072-8/+64
| |/
* | Gaussian normal distribution of PRNG via Box-Muller transformIvar Flakstad2024-01-071-3/+28
* | Implement hybrid Tausworthe + LCG psuedo random number generator in metalIvar Flakstad2024-01-051-0/+41
|/
* Sketch the minimal mamba example. (#1465)Laurent Mazare2023-12-221-1/+0
* Optimizing decode matmul (Phi at 28tok/s on M3).Nicolas Patry2023-12-201-0/+43