o3 Vision: pricing, context window and benchmarks
o3 Vision is a multimodal model from OpenAI. It offers a 1,000,000-token context window, costs $15.24 per million input tokens and $65.25 per million output tokens, and averages 81.8 across the index's benchmark suite.
Pricing
o3 Vision costs $15.24 per million input tokens and $65.25 per million output tokens. A blended 3:1 input-to-output mix works out to about $27.74 per million tokens. Cached input is listed at $1.47 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $15.24 | per 1M tokens |
| Output | $65.25 | per 1M tokens |
| Cached input | $1.47 | per 1M tokens |
| Blended 3:1 | $27.74 | estimate per 1M tokens |
| 1M in + 1M out | $80.49 | 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 | 83.6 | 92 | 85th | |
| GPQA Diamond | 77.8 | 92 | 85th | |
| SWE-bench Verified | 71.8 | 79 | 87th | |
| HumanEval | 85.6 | 104 | 83th | |
| MATH-500 | 84.4 | 118 | 80th | |
| AIME 2025 | 82.4 | 62 | 90th | |
| LiveCodeBench | 73.7 | 117 | 81th | |
| IFEval | 89.3 | 78 | 87th | |
| TruthfulQA | 85.8 | 55 | 91th | |
| HellaSwag | 89.7 | 96 | 84th | |
| MMMU | 75.7 | 147 | 76th | |
| Arena Elo | 1436.5 | 104 | 83th |
What o3 Vision is good for
o3 Vision 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 $15.24 per million input tokens, it sits in the mid-range of the index, and the 1,000,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 1,000,000-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 $15.24 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| GPT-4.1 Vision | OpenAI | 1000000 | $14.88 | 82.0 |
| Gemini 2.5 Pro Vision | Google DeepMind | 200000 | $14.91 | 82.0 |
| Gemini 3 Flash Vision | Google DeepMind | 400000 | $15.06 | 81.9 |
| 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 |
FAQ
How much does o3 Vision cost?
o3 Vision is listed at $15.24 per million input tokens and $65.25 per million output tokens. Cached input is $1.47 per million tokens. A blended 3:1 input-to-output mix is about $27.74 per million tokens.
What is the context window of o3 Vision?
o3 Vision has a context window of 1,000,000 tokens and can emit up to 8,192 output tokens per request in this index.
Is o3 Vision open weights?
No. o3 Vision is a proprietary model with a proprietary licence, so you cannot download the weights.
How does o3 Vision score on benchmarks?
o3 Vision averages 81.8 across the scored benchmarks in this index, ranking #88 of 600 models. Its strongest relative result is Arena Elo at 1436.5.
What is o3 Vision best for?
o3 Vision 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 $15.24 per million input tokens, it sits in the mid-range of the index, and the 1,000,000-token context window is enough for large files and multi-file diffs.