huggingface/candle Issues
Quantized-t5 models on Cuda
UpdatedImplement torch.scatter
Updated 1Implement unfold function
UpdatedONNX: MaxPool with pads != 0
UpdatedUnsupported op_type Pad for op
Closed 1Tensor::to_scalar very high latency
Updated 5Meta voice WASM example?
Updated 1SeparableConv2d implementation
Updated 1Implement `torch.bucketize`
Closed 2Quantization issue - Mixtral 8x22b
Updated 2How to slice a tensor?
Closed 1VarBuilder::from_bytes?
Closed 6Model to architecture mapping
Closed 2