family llama · 9 variants · unknown
| benchmark | category | best score |
|---|---|---|
| GPQA | reasoning | 89.0 |
| IFEval | instruction-following | 73.29 |
| MMLU-Pro | knowledge | 27.81 |
| BBH | reasoning | 27.13 |
| MATH | math | 6.42 |
| MUSR | reasoning | 3.21 |
| variant | kind | quant | bits | vRAM | dl | org | year |
|---|---|---|---|---|---|---|---|
| Meta-Llama-3.1-8B-Instruct-GGUF | gguf | gguf | 2.0 | 286,108 | bartowski | 2024 | |
| Meta-Llama-3.1-8B-Instruct-GGUF | gguf | gguf | 4.0 | 72,085 | lmstudio-community | 2024 | |
| Meta-Llama-3.1-8B-Instruct-abliterated-GGUF | gguf | gguf | 2.0 | 33,844 | mlabonne | 2024 | |
| Meta-Llama-3.1-8B-Instruct-abliterated | abliterated | abliterated | 24,003 | mlabonne | 2024 | ||
| Llama-3.1-8B-Instruct-GGUF | gguf | gguf | 16.0 | 17,812 | unsloth | 2025 | |
| Llama3.1-8B-Chinese-Chat | gguf | gguf | 16.0 | 7,534 | shenzhi-wang | 2024 | |
| Meta-Llama-3.1-8B-Instruct-Q4_0-GGUF | gguf | gguf | 4.0 | 2,310 | ggml-org | 2024 | |
| cogvlm2-llama3-caption | original | bf16 | 16.0 | 877 | zai-org | 2025 | |
| Meta-Llama-3.1-8B-Instruct-BNB-NF4 | bnb-4bit | 4bit | 368 | hugging-quants | 2024 |