GPT-5.2 Vision: pricing, context window and benchmarks
GPT-5.2 Vision is a frontier model from OpenAI. It offers a 1,000,000-token context window, costs $19.03 per million input tokens and $85.72 per million output tokens, and averages 87.4 across the index's benchmark suite.
Pricing
GPT-5.2 Vision costs $19.03 per million input tokens and $85.72 per million output tokens. A blended 3:1 input-to-output mix works out to about $35.70 per million tokens. Cached input is listed at $1.83 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $19.03 | per 1M tokens |
| Output | $85.72 | per 1M tokens |
| Cached input | $1.83 | per 1M tokens |
| Blended 3:1 | $35.70 | estimate per 1M tokens |
| 1M in + 1M out | $104.75 | 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 | 89.5 | 9 | 99th | |
| GPQA Diamond | 82.1 | 28 | 95th | |
| SWE-bench Verified | 78.8 | 11 | 98th | |
| HumanEval | 91.2 | 31 | 95th | |
| MATH-500 | 92.9 | 12 | 98th | |
| AIME 2025 | 85.7 | 28 | 95th | |
| LiveCodeBench | 83.4 | 11 | 98th | |
| IFEval | 93.9 | 12 | 98th | |
| TruthfulQA | 85.3 | 63 | 90th | |
| HellaSwag | 94.6 | 7 | 99th | |
| MMMU | 84.4 | 10 | 98th | |
| Arena Elo | 1484.3 | 18 | 97th |
What GPT-5.2 Vision is good for
GPT-5.2 Vision 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 $19.03 per million input tokens and a 1,000,000-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 (87.4), placing it #18 of 600.
- 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 $19.03 per million tokens.
- Cost can climb quickly on long outputs and high-volume traffic.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| GPT-5.2 Reasoning | OpenAI | 1000000 | $20.98 | 87.5 |
| Gemini 3 Pro Reasoning | Google DeepMind | 200000 | $19.69 | 87.5 |
| o4 Reasoning | OpenAI | 1000000 | $19.69 | 87.5 |
| GPT-5.2 Turbo | OpenAI | 1000000 | $20.73 | 87.4 |
| GPT-5.2 Vision | OpenAI | 1000000 | $19.03 | 87.4 |
| o4 Pro | OpenAI | 1000000 | $19.37 | 87.3 |
FAQ
How much does GPT-5.2 Vision cost?
GPT-5.2 Vision is listed at $19.03 per million input tokens and $85.72 per million output tokens. Cached input is $1.83 per million tokens. A blended 3:1 input-to-output mix is about $35.70 per million tokens.
What is the context window of GPT-5.2 Vision?
GPT-5.2 Vision has a context window of 1,000,000 tokens and can emit up to 16,384 output tokens per request in this index.
Is GPT-5.2 Vision open weights?
No. GPT-5.2 Vision is a proprietary model with a proprietary licence, so you cannot download the weights.
How does GPT-5.2 Vision score on benchmarks?
GPT-5.2 Vision averages 87.4 across the scored benchmarks in this index, ranking #18 of 600 models. Its strongest relative result is Arena Elo at 1484.3.
What is GPT-5.2 Vision best for?
GPT-5.2 Vision 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 $19.03 per million input tokens and a 1,000,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.