GPQA Diamond
?what it ishighest score on this metric 96.4%
36º / 198 · Epoch AI, latest version we checked
Benchmarks, price, context and versions
MiniMax M3, MiniMax, minimax m3.1 flash, minimax m3.1, minimax m3 reddit
Direct answer: MiniMax M3 is a model from MiniMax. 90.9% on GPQA Diamond, per Epoch AI. On OpenRouter, US$ 0.30 per 1 million input tokens and US$ 1.20 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%
36º / 198 · Epoch AI, latest version we checked
highest score on this metric 100%
99º / 186 · Epoch AI, latest version we checked
highest score on this metric 96.5%
26º / 135 · OpenRouter Benchmarks, latest version we checked
highest score on this metric 82%
38º / 131 · OpenRouter Benchmarks, 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$ 0.30 per 1 million input tokens and US$ 1.20 per 1 million output tokens.
MiniMax M3 is a model from MiniMax, with open weights, used for text, agents.
1,048,576 tokens, per OpenRouter.
Yes, it has open weights. License: no data.
90.9% on GPQA Diamond, per Epoch AI.
No data: there is no SWE-bench Verified measurement for it in open sources.
It was made by MiniMax, a lab from China.
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.