Gemini 3 Flash Reasoning: pricing, context window and benchmarks
Gemini 3 Flash Reasoning is a frontier model from Google DeepMind. It offers a 128,000-token context window, costs $17.05 per million input tokens and $66.77 per million output tokens, and averages 83.7 across the index's benchmark suite.
Pricing
Gemini 3 Flash Reasoning costs $17.05 per million input tokens and $66.77 per million output tokens. A blended 3:1 input-to-output mix works out to about $29.48 per million tokens. Cached input is listed at $1.41 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $17.05 | per 1M tokens |
| Output | $66.77 | per 1M tokens |
| Cached input | $1.41 | per 1M tokens |
| Blended 3:1 | $29.48 | estimate per 1M tokens |
| 1M in + 1M out | $83.82 | 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 | 87.2 | 32 | 95th | |
| GPQA Diamond | 81.8 | 32 | 95th | |
| SWE-bench Verified | 72.5 | 67 | 89th | |
| HumanEval | 86.9 | 91 | 85th | |
| MATH-500 | 86.2 | 89 | 85th | |
| AIME 2025 | 79.4 | 96 | 84th | |
| LiveCodeBench | 77.7 | 69 | 89th | |
| IFEval | 90.2 | 67 | 89th | |
| TruthfulQA | 86.6 | 43 | 93th | |
| HellaSwag | 92.2 | 40 | 93th | |
| MMMU | 79.8 | 58 | 90th | |
| Arena Elo | 1446.9 | 85 | 86th |
What Gemini 3 Flash Reasoning is good for
Gemini 3 Flash Reasoning sits in the frontier tier of the index. It is aimed at work where answer quality matters more than unit cost: hard reasoning, multi-step coding, document analysis and multimodal inputs. At $17.05 per million input tokens and a 128,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.
Strengths
- Top-tier average benchmark score (83.7), placing it #67 of 600.
- Large 128,000-token context window for long documents and agent traces.
- Accepts more than text, including image input where the provider supports it.
- Reasoning-oriented profile, useful for multi-step problems and structured analysis.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $17.05 per million tokens.
- Cost can climb quickly on long outputs and high-volume traffic.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Claude Haiku 4.5 Max | Anthropic | 200000 | $15.98 | 83.8 |
| Grok 3 Pro | xAI | 400000 | $16.61 | 83.8 |
| Claude 3.7 Sonnet Max | Anthropic | 128000 | $16.88 | 83.7 |
| Claude 3.7 Sonnet Reasoning | Anthropic | 128000 | $16.30 | 83.7 |
| DeepSeek-V3.2 Instruct | DeepSeek | 128000 | $5.91 | 83.7 |
| Gemini 3 Flash Reasoning | Google DeepMind | 128000 | $17.05 | 83.7 |
FAQ
How much does Gemini 3 Flash Reasoning cost?
Gemini 3 Flash Reasoning is listed at $17.05 per million input tokens and $66.77 per million output tokens. Cached input is $1.41 per million tokens. A blended 3:1 input-to-output mix is about $29.48 per million tokens.
What is the context window of Gemini 3 Flash Reasoning?
Gemini 3 Flash Reasoning has a context window of 128,000 tokens and can emit up to 32,768 output tokens per request in this index.
Is Gemini 3 Flash Reasoning open weights?
No. Gemini 3 Flash Reasoning is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Gemini 3 Flash Reasoning score on benchmarks?
Gemini 3 Flash Reasoning averages 83.7 across the scored benchmarks in this index, ranking #67 of 600 models. Its strongest relative result is Arena Elo at 1446.9.
What is Gemini 3 Flash Reasoning best for?
Gemini 3 Flash Reasoning sits in the frontier tier of the index. It is aimed at work where answer quality matters more than unit cost: hard reasoning, multi-step coding, document analysis and multimodal inputs. At $17.05 per million input tokens and a 128,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.