Llama 4 Maverick Instruct: pricing, context window and benchmarks
Llama 4 Maverick Instruct is a multimodal model from Meta AI. It offers a 200,000-token context window, costs $5.33 per million input tokens and $23.04 per million output tokens, and averages 81.7 across the index's benchmark suite.
Pricing
Llama 4 Maverick Instruct costs $5.33 per million input tokens and $23.04 per million output tokens. A blended 3:1 input-to-output mix works out to about $9.76 per million tokens. Cached input is listed at $0.63 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $5.33 | per 1M tokens |
| Output | $23.04 | per 1M tokens |
| Cached input | $0.63 | per 1M tokens |
| Blended 3:1 | $9.76 | estimate per 1M tokens |
| 1M in + 1M out | $28.37 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 84.6 | 70 | 88th | |
| GPQA Diamond | 79.5 | 67 | 89th | |
| SWE-bench Verified | 69.1 | 130 | 78th | |
| HumanEval | 85.0 | 117 | 81th | |
| MATH-500 | 83.4 | 139 | 77th | |
| AIME 2025 | 79.9 | 91 | 85th | |
| LiveCodeBench | 75.8 | 84 | 86th | |
| IFEval | 88.9 | 86 | 86th | |
| TruthfulQA | 84.0 | 84 | 86th | |
| HellaSwag | 89.6 | 98 | 84th | |
| MMMU | 79.0 | 73 | 88th | |
| Arena Elo | 1438.5 | 100 | 83th |
What Llama 4 Maverick Instruct is good for
Llama 4 Maverick Instruct is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $5.33 per million input tokens, it sits in the mid-range of the index, and the 200,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 200,000-token context window for long documents and agent traces.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- No standout weakness in the indexed specifications, but real-world behaviour still needs testing.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| o3 Vision | OpenAI | 1000000 | $15.24 | 81.8 |
| Llama 4 Maverick Instruct | Meta AI | 200000 | $5.33 | 81.7 |
| Mistral Large 3 | Mistral AI | 128000 | $5.41 | 81.7 |
| Claude 3.7 Sonnet | Anthropic | 200000 | $14.91 | 81.6 |
| Gemini 2.5 Pro | Google DeepMind | 200000 | $15.20 | 81.6 |
| Llama 4 Maverick | Meta AI | 200000 | $5.36 | 81.6 |
FAQ
How much does Llama 4 Maverick Instruct cost?
Llama 4 Maverick Instruct is listed at $5.33 per million input tokens and $23.04 per million output tokens. Cached input is $0.63 per million tokens. A blended 3:1 input-to-output mix is about $9.76 per million tokens.
What is the context window of Llama 4 Maverick Instruct?
Llama 4 Maverick Instruct has a context window of 200,000 tokens and can emit up to 16,384 output tokens per request in this index.
Is Llama 4 Maverick Instruct open weights?
Yes. Llama 4 Maverick Instruct is listed as an open-weight model under a open weights licence.
How does Llama 4 Maverick Instruct score on benchmarks?
Llama 4 Maverick Instruct averages 81.7 across the scored benchmarks in this index, ranking #89 of 600 models. Its strongest relative result is Arena Elo at 1438.5.
What is Llama 4 Maverick Instruct best for?
Llama 4 Maverick Instruct is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $5.33 per million input tokens, it sits in the mid-range of the index, and the 200,000-token context window is enough for large files and multi-file diffs.