Meta/Llama 4/Apr 5, 2025
Llama 4 Scout
The long-context specialist in the Llama 4 family. A 17B active parameter model with 16 experts and an industry-leading 10M-token context window. Beats Gemma 3, Gemini 2.0 Flash-Lite, and Mistral 3.1 across a broad range of benchmarks.
textcodereasoningvision
Arena ELO
1,322
Input Price
$0.18/M
Output Price
$0.66/M
Speed
149 t/s
Context
10M
Latency
780ms
Capability Assessment
GPQA Diamond57.2%
MMMU Pro69.4%
Comparative Analysis
| Metric | Llama 4 Scout | Claude Opus 4.6 | Gemini 3.1 Pro | Grok 4 |
|---|---|---|---|---|
| SWE-bench | — | 80.8% | 80.6% | 72.0% |
| AIME 2025 | — | 100.0% | 91.2% | 94.0% |
| GPQA Diamond | 57.2% | 91.3% | 94.3% | 88.0% |
| MMLU | 87.2% | 91.0% | 92.6% | — |
| Input $/M | $0.18 | $5 | $2 | $3 |
| Output $/M | $0.66 | $25 | $12 | $15 |