GPQA Diamond
?what it ishighest score on this metric 96.4%
32º / 198 · Epoch AI, latest version we checked
Benchmarks, price, context and versions
GPT 5.2, gpt-5.2 pricing, gpt-5.2-codex, gpt-5.2-chat-latest
Direct answer: GPT-5.2 is a model from OpenAI. 91.4% on GPQA Diamond, per Epoch AI. On OpenRouter, US$ 1.75 per 1 million input tokens and US$ 14.00 per 1 million output tokens.
This block follows a rule of this section, using source data: if the model is in the top 25% of the metric, the item goes under "Good for"; if it is in the bottom half, it goes under "Not the best for". The middle band appears in neither. Price and context use OpenRouter; scores use Epoch AI or OpenRouter Benchmarks, in the latest version we checked.
highest score on this metric 96.4%
32º / 198 · Epoch AI, latest version we checked
highest score on this metric 100%
37º / 186 · Epoch AI, latest version we checked
highest score on this metric 95.1%
34º / 86 · Epoch AI, latest version we checked
highest score on this metric 94%
53º / 87 · Epoch AI, latest version we checked
highest score on this metric 83.5%
17º / 32 · Epoch AI, latest version we checked
Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch.
On OpenRouter, US$ 1.75 per 1 million input tokens and US$ 14.00 per 1 million output tokens.
400,000 tokens, per OpenRouter.
It scores 73.8% on SWE-bench Verified, per Epoch AI.
GPT-5.2 is a model from OpenAI, with closed weights, used for text, agents, code.
No. It is available through API only.
91.4% on GPQA Diamond, per Epoch AI.
It was made by OpenAI, a lab from the US.
provisional data Questions ordered by what shows up in public Google search suggestions (no volume). Answers use source data only.
Dates come from the sources: OpenRouter listing or Hugging Face repo creation, which may differ from the official launch. Family and version were grouped by rule from the name.