Laurent Mazare b8263aa15c Quantized support for f16 and f32 (#457) ...

* Add f32 as a quantized type.

* Add f16 as a quantized type too.

2023-08-15 21:09:37 +01:00

..

bert

Add a cuda kernel for upsampling. (#441)

2023-08-14 13:12:17 +01:00

bigcode

Add a cuda kernel for upsampling. (#441)

2023-08-14 13:12:17 +01:00

custom-ops

Use bail rather than wrapping a string where possible. (#249)

2023-07-26 15:42:46 +01:00

falcon

Add a cuda kernel for upsampling. (#441)

2023-08-14 13:12:17 +01:00

ggml

Quantized support for f16 and f32 (#457)

2023-08-15 21:09:37 +01:00

llama

Tweak the llama example. (#450)

2023-08-15 12:18:20 +01:00

llama2-c

Support the Accelerate BLAS on macOS. (#325)

2023-08-05 17:25:24 +01:00

llama_multiprocess

Remove the checkpoint conversion script. (#405)

2023-08-11 05:59:48 +01:00

mnist-training

Add the candle-datasets crate (#322)

2023-08-05 08:56:50 +01:00

musicgen

Remove the embedding ops in favor of index-select. (#299)

2023-08-02 05:42:11 +01:00

stable-diffusion

Track the conv2d operations in stable-diffusion. (#431)

2023-08-13 15:58:26 +01:00

whisper

Add a cuda kernel for upsampling. (#441)

2023-08-14 13:12:17 +01:00