File size: 1,558 Bytes
19ea320 90c91dc 19ea320 90c91dc 19ea320 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 |
---
tags:
- fp8
- vllm
---
# Mixtral-8x22B-Instruct-v0.1-FP8
## Model Overview
Mixtral-8x22B-Instruct-v0.1 quantized to FP8 weights and activations using per-tensor quantization, ready for inference with vLLM >= 0.5.0.
## Usage and Creation
Produced using [AutoFP8 with calibration samples from ultrachat](https://github.com/neuralmagic/AutoFP8/blob/147fa4d9e1a90ef8a93f96fc7d9c33056ddc017a/example_dataset.py).
## Evaluation
### Open LLM Leaderboard evaluation scores
| | Mixtral-8x22B-Instruct-v0.1 | Mixtral-8x22B-Instruct-v0.1-FP8<br>(this model) |
| :------------------: | :----------------------: | :------------------------------------------------: |
| arc-c<br>25-shot | 72.70 | 69.19 |
| hellaswag<br>10-shot | 89.08 | 82.49 |
| mmlu<br>5-shot | 77.77 | 70.61 |
| truthfulqa<br>0-shot | 68.14 | 65.73 |
| winogrande<br>5-shot | 85.16 | 82.63 |
| gsm8k<br>5-shot | 82.03 | 76.57 |
| **Average<br>Accuracy** | **79.15** | **74.53** |
| **Recovery** | **100%** | **94.17%** |
|