Quantized Model Information
This repository is a 'FP8-Dynamic' quantized version of
meta-llama/Llama-3.3-70B-Instruct, originally released by Meta AI.
For usage instructions please refer to the original model meta-llama/Llama-3.3-70B-Instruct.
Performance
All benchmarks were done using the LLM Evaluation Harness
| Llama-3.3-70B-Instruct-FP8-Dynamic | Llama-3.3-70B-Instruct (base) | recovery | ||
|---|---|---|---|---|
| mmlu | - | xx | xx | xx |
| xx | xx | xx | ||
| hellaswag | acc | 65.69 | - | |
| acc_sterr | 0.47 | - | ||
| acc_norm | 84.36 | - | ||
| acc_sterr | 0.36 | - |
- Downloads last month
- 6
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for just-add-ai/Llama-3.3-70B-Instruct-FP8-Dynamic
Base model
meta-llama/Llama-3.1-70B
Finetuned
meta-llama/Llama-3.3-70B-Instruct