M
Minimax
MiniMax M2.5 (free)
MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1...
- Input / 1M tokens
- Free
- Output / 1M tokens
- Free
- Context window
- 197K tokens
- Provider
- Minimax
Performance
Median streaming throughput and first-token latency measured by Artificial Analysis.
- Output tokens / sec
- 61 t/s
- Time to first token
- 1.80s
Benchmarks
Intelligence, coding, and math indexes plus the underlying evaluation scores.
- Intelligence Index
- 42
- Coding Index
- 37
- Math Index
- —
- MMLU-Pro
- —
- GPQA
- 84.8%
- HLE
- 19.1%
- LiveCodeBench
- —
- SciCode
- 42.6%
- MATH-500
- —
- AIME
- —
Benchmarks via Artificial Analysis