meta-llama / llama-3.1 family / text-generation
Llama-3.1-8B-Instruct
meta-llama/Llama-3.1-8B-Instruct
Benchmarks
Where it scores.
General Provisional
51.7
Coding Provisional
45.3
Math Provisional
45.3
Reasoning Provisional
45.3
Normalized 0–100 across indexed models.
Runnable variants
Pick a size.
| Precision | Format | Size | Params | File |
|---|---|---|---|---|
| FP16 / smallest | SAFETENSORS | 16.1 GB | — | model-00001-of-00004.safetensors |
Sizes are file sizes, not peak runtime memory.
Raw benchmarks
The inputs.
| Source | Task | Raw | Normalized |
|---|---|---|---|
| Arena | ArenaScore | 1211.2 | 63.9 |
| ZeroEval | GPQA | 30.4 | 30.4 |
From the Hugging Face model card
About this model.
From the Hugging Face model card