Fusion Reasoning: pricing, context window and benchmarks
Fusion Reasoning is a mid-range model from OpenRouter. It offers a 65,536-token context window, costs $8.59 per million input tokens and $32.34 per million output tokens, and averages 72.2 across the index's benchmark suite.
Pricing
Fusion Reasoning costs $8.59 per million input tokens and $32.34 per million output tokens. A blended 3:1 input-to-output mix works out to about $14.53 per million tokens. Cached input is listed at $0.95 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $8.59 | per 1M tokens |
| Output | $32.34 | per 1M tokens |
| Cached input | $0.95 | per 1M tokens |
| Blended 3:1 | $14.53 | estimate per 1M tokens |
| 1M in + 1M out | $40.93 | 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 | 73.3 | 365 | 39th | |
| GPQA Diamond | 67.4 | 323 | 46th | |
| SWE-bench Verified | 61.7 | 288 | 52th | |
| HumanEval | 78.2 | 285 | 53th | |
| MATH-500 | 73.1 | 356 | 41th | |
| AIME 2025 | 63.3 | 364 | 39th | |
| LiveCodeBench | 66.6 | 259 | 57th | |
| IFEval | 82.0 | 278 | 54th | |
| TruthfulQA | 73.8 | 358 | 40th | |
| HellaSwag | 84.1 | 274 | 54th | |
| MMMU | 71.0 | 272 | 55th | |
| Arena Elo | 1357.6 | 348 | 42th |
What Fusion Reasoning is good for
Fusion Reasoning is a general-purpose model in the index. It handles text generation, summarization and question answering, and it can be a sensible default when you need a balance of quality, price and context. At $8.59 per million input tokens with a 65,536-token window, check its benchmark profile against the specific task before committing.
Strengths
- Accepts more than text, including image input where the provider supports it.
- Reasoning-oriented profile, useful for multi-step problems and structured analysis.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $8.59 per million tokens.
- Average benchmark score is below the index median; test it on your own task.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Firefunction | Fireworks AI | 16384 | $8.16 | 72.3 |
| Fusion Pro | OpenRouter | 65536 | $8.69 | 72.2 |
| Fusion Reasoning | OpenRouter | 65536 | $8.59 | 72.2 |
| Llama 3 Modal Base | Modal | 16384 | $3.21 | 72.2 |
| Llama 3.1 Base | Meta AI | 32768 | $3.39 | 72.2 |
| Seed-OSS Base | ByteDance | 16384 | $3.14 | 72.2 |
FAQ
How much does Fusion Reasoning cost?
Fusion Reasoning is listed at $8.59 per million input tokens and $32.34 per million output tokens. Cached input is $0.95 per million tokens. A blended 3:1 input-to-output mix is about $14.53 per million tokens.
What is the context window of Fusion Reasoning?
Fusion Reasoning has a context window of 65,536 tokens and can emit up to 65,536 output tokens per request in this index.
Is Fusion Reasoning open weights?
No. Fusion Reasoning is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Fusion Reasoning score on benchmarks?
Fusion Reasoning averages 72.2 across the scored benchmarks in this index, ranking #322 of 600 models. Its strongest relative result is Arena Elo at 1357.6.
What is Fusion Reasoning best for?
Fusion Reasoning is a general-purpose model in the index. It handles text generation, summarization and question answering, and it can be a sensible default when you need a balance of quality, price and context. At $8.59 per million input tokens with a 65,536-token window, check its benchmark profile against the specific task before committing.