Meta
Llama 4 Maverick
Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward...
Introducing Llama 4Model Snapshot
ReleaseApr 5, 2025
Context1,048,576
Overall
Rank #290 across 566
14.3
Coding
Rank #177 across 222
16.3
Blended Price
Rank #111 across 371
$0.475/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Output Speed98.3 tok/s
First Token0.62s
Max Output16,384
Pricing
Blended$0.475/M
Input$0.350/M
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 14.3 | #290 |
| Artificial Analysis Coding Index | coding | 16.3 | #177 |
| Artificial Analysis Math Index | math | 19.3 | #211 |
| MMLU-Pro | reasoning | 80.9% | #99 |