Llama 4 Maverick Base: pricing, context window and benchmarks
Llama 4 Maverick Base is a multimodal model from Meta AI. It offers a 128,000-token context window, costs $4.68 per million input tokens and $21.03 per million output tokens, and averages 79.1 across the index's benchmark suite.
Pricing
Llama 4 Maverick Base costs $4.68 per million input tokens and $21.03 per million output tokens. A blended 3:1 input-to-output mix works out to about $8.77 per million tokens. Cached input is listed at $0.62 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $4.68 | per 1M tokens |
| Output | $21.03 | per 1M tokens |
| Cached input | $0.62 | per 1M tokens |
| Blended 3:1 | $8.77 | estimate per 1M tokens |
| 1M in + 1M out | $25.71 | 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 | 83.3 | 100 | 83th | |
| GPQA Diamond | 71.8 | 224 | 63th | |
| SWE-bench Verified | 67.9 | 148 | 75th | |
| HumanEval | 85.4 | 107 | 82th | |
| MATH-500 | 83.3 | 142 | 76th | |
| AIME 2025 | 77.8 | 117 | 81th | |
| LiveCodeBench | 69.3 | 203 | 66th | |
| IFEval | 87.4 | 122 | 80th | |
| TruthfulQA | 81.2 | 150 | 75th | |
| HellaSwag | 89.2 | 109 | 82th | |
| MMMU | 73.8 | 197 | 67th | |
| Arena Elo | 1425.2 | 129 | 79th |
What Llama 4 Maverick Base is good for
Llama 4 Maverick Base 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 $4.68 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 128,000-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 |
|---|---|---|---|---|
| Llama 4 Maverick Base | Meta AI | 128000 | $4.68 | 79.1 |
| Llama 4 Scout Instruct | Meta AI | 32768 | $4.59 | 79.1 |
| Mistral Large 3 Chat | Mistral AI | 1000000 | $4.90 | 79.1 |
| Qwen3 Chat | Alibaba Cloud | 128000 | $4.75 | 79.1 |
| Sonar Pro | Perplexity | 128000 | $13.01 | 79.1 |
| Claude Opus 4.5 Mini | Anthropic | 65536 | $12.72 | 79.0 |
FAQ
How much does Llama 4 Maverick Base cost?
Llama 4 Maverick Base is listed at $4.68 per million input tokens and $21.03 per million output tokens. Cached input is $0.62 per million tokens. A blended 3:1 input-to-output mix is about $8.77 per million tokens.
What is the context window of Llama 4 Maverick Base?
Llama 4 Maverick Base has a context window of 128,000 tokens and can emit up to 128,000 output tokens per request in this index.
Is Llama 4 Maverick Base open weights?
Yes. Llama 4 Maverick Base is listed as an open-weight model under a open weights licence.
How does Llama 4 Maverick Base score on benchmarks?
Llama 4 Maverick Base averages 79.1 across the scored benchmarks in this index, ranking #144 of 600 models. Its strongest relative result is Arena Elo at 1425.2.
What is Llama 4 Maverick Base best for?
Llama 4 Maverick Base 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 $4.68 per million input tokens, it sits in the mid-range of the index, and the 128,000-token context window is enough for large files and multi-file diffs.