Hugging Face's logo

shisa-ai
/
shisa-v2-llama3.1-405b-W8A8-INT8

Text Generation
Transformers
Safetensors
llama
conversational
text-generation-inference
8-bit precision
compressed-tensors
Model card Files Community

Quantized w/ llm-compressor using shisa-ai/shisa-v2-sharegpt as the calibration set.

Original model: shisa-ai/shisa-v2-llama3.1-405b

Downloads last month
6
Safetensors
Model size
406B params
Tensor type
BF16
·
I8
·
Inference Providers NEW
Text Generation
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for shisa-ai/shisa-v2-llama3.1-405b-W8A8-INT8

Base model

meta-llama/Llama-3.1-405B
Finetuned
meta-llama/Llama-3.1-405B-Instruct
Finetuned
shisa-ai/shisa-v2-llama3.1-405b
Quantized
(4)
this model

Datasets used to train shisa-ai/shisa-v2-llama3.1-405b-W8A8-INT8

Collection including shisa-ai/shisa-v2-llama3.1-405b-W8A8-INT8