Llama 3.1: pricing, context window and benchmarks
Llama 3.1 is a multimodal model from Meta AI. It offers a 200,000-token context window, costs $4.16 per million input tokens and $15.79 per million output tokens, and averages 76.3 across the index's benchmark suite.
Pricing
Llama 3.1 costs $4.16 per million input tokens and $15.79 per million output tokens. A blended 3:1 input-to-output mix works out to about $7.07 per million tokens. Cached input is listed at $0.60 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $4.16 | per 1M tokens |
| Output | $15.79 | per 1M tokens |
| Cached input | $0.60 | per 1M tokens |
| Blended 3:1 | $7.07 | estimate per 1M tokens |
| 1M in + 1M out | $19.95 | 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 | 78.8 | 220 | 63th | |
| GPQA Diamond | 69.8 | 272 | 55th | |
| SWE-bench Verified | 67.9 | 147 | 76th | |
| HumanEval | 81.8 | 189 | 69th | |
| MATH-500 | 82.8 | 150 | 75th | |
| AIME 2025 | 69.5 | 268 | 55th | |
| LiveCodeBench | 66.7 | 258 | 57th | |
| IFEval | 82.3 | 267 | 56th | |
| TruthfulQA | 78.4 | 240 | 60th | |
| HellaSwag | 87.2 | 166 | 72th | |
| MMMU | 74.2 | 180 | 70th | |
| Arena Elo | 1407.6 | 187 | 69th |
What Llama 3.1 is good for
Llama 3.1 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 $4.16 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 |
|---|---|---|---|---|
| Command R+ | Cohere | 32768 | $3.96 | 76.3 |
| Llama 3.1 | Meta AI | 200000 | $4.16 | 76.3 |
| Command R+ Instruct | Cohere | 200000 | $4.16 | 76.2 |
| Llama 3 Baseten Instruct | Baseten | 200000 | $3.91 | 76.2 |
| Llama 3 Hyperbolic Instruct | Hyperbolic | 200000 | $3.79 | 76.2 |
| o3 Mini | OpenAI | 128000 | $10.68 | 76.2 |
FAQ
How much does Llama 3.1 cost?
Llama 3.1 is listed at $4.16 per million input tokens and $15.79 per million output tokens. Cached input is $0.60 per million tokens. A blended 3:1 input-to-output mix is about $7.07 per million tokens.
What is the context window of Llama 3.1?
Llama 3.1 has a context window of 200,000 tokens and can emit up to 32,768 output tokens per request in this index.
Is Llama 3.1 open weights?
Yes. Llama 3.1 is listed as an open-weight model under a open weights licence.
How does Llama 3.1 score on benchmarks?
Llama 3.1 averages 76.3 across the scored benchmarks in this index, ranking #212 of 600 models. Its strongest relative result is Arena Elo at 1407.6.
What is Llama 3.1 best for?
Llama 3.1 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 $4.16 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.