Llama 4
Llama-4-Scout-17B-16E-Instruct
Meta AI · open-restricted llama4
| Parameters (B) | 108.642 |
|---|---|
| Architecture | moe |
| Layers / KV heads / head dim | 48 / 8 / 128 |
| Context window | 10485760 |
| Modalities | in: text, image; out: text |
| GGUF | available |
| Commercial use | yes |
Minimum hardware (estimate)
At Q4_K_M and 8k context, total working set ≈ 68.1 GiB. On a machine with 64 GiB system RAM and no discrete GPU: Won't run. Limiting factor: No GPU and system RAM insufficient for weights + KV.
Scores
| Benchmark | Value | Uncertainty | Run by | Source |
|---|---|---|---|---|
| GPQA Diamond | 51.8% | unknown | third-party | epoch-ai |
| MATH Level 5 | 62.3% | unknown | third-party | epoch-ai |
| OTIS Mock AIME 2024-2025 | 7.8% | unknown | third-party | epoch-ai |
| FrontierMath | 0.0% | unknown | third-party | epoch-ai |
Provenance
- identity, license, architecture, parameter count: https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E-Instruct (retrieved 2026-08-10)
HF: meta-llama/Llama-4-Scout-17B-16E-Instruct · snapshot 2026-08-10