jbochi's picture
Add instructions to quantize model
e40158b
|
raw
history blame
511 Bytes
---
license: apache-2.0
datasets:
- grammarly/coedit
language:
- en
tags:
- text-generation-inference
- candle
---
Quantized weights of [coedit](https://github.com/vipulraheja/coedit) for inference with [candle](https://github.com/huggingface/candle/tree/main/candle-examples/examples/quantized-t5).
Conversion command, using candle:
```shell
cargo run --example tensor-tools --release -- quantize \
--quantization q6k \
/path/to/coedit-<version>/model.safetensors \
--out-file model<version>.gguf
```