turboderp's picture
Update README.md
e5ef4dc
|
raw
history blame
No virus
1.09 kB
EXL2 quants of Llama2-70B-chat
- [2.30 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/2.3bpw)
- [2.35 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/2.35bpw)
- [2.40 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/2.4bpw)
- [2.45 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/2.45bpw)
- [2.50 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/2.5bpw)
- [2.55 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/2.55bpw)
- [2.60 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/2.6bpw)
- [2.70 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/2.7bpw)
- [3.00 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/3.0bpw)
- [4.00 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/4.0bpw)
- [4.65 bits per weight](https://huggingface.co/turboderp/Llama2-70B-exl2/tree/4.65bpw)
[measurement.json](https://huggingface.co/turboderp/Llama2-70B-exl2/blob/main/measurement.json)