Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
comaniac
/
Mixtral-8x7B-Instruct-v0.1-FP8-v2
like
0
Text Generation
Transformers
Safetensors
mixtral
conversational
text-generation-inference
Inference Endpoints
fp8
Model card
Files
Files and versions
Community
Train
Deploy
Use this model
main
Mixtral-8x7B-Instruct-v0.1-FP8-v2
/
README.md
comaniac
Update README.md
200f867
verified
3 months ago
preview
code
|
raw
Copy download link
history
blame
contribute
delete
No virus
231 Bytes
Mixtral-8x7B-Instruct-v0.1-FP8-v2
Weights and activations are per-tensor quantized to float8_e4m3.
Quantization with AutoFP8.
Calibration dataset: Ultrachat-200k
Samples: 4096
Sequence length: 8192
Evaluation
TBA