Side by side
See the difference.
Scores, capabilities and costs for the models you’re considering.
Scroll to compare models →
| 3 models | minimaxMiniMax: MiniMax M2.5Remove × | minimaxMiniMax: MiniMax M2Remove × | minimaxMiniMax: MiniMax M3Remove × |
|---|---|---|---|
| What’s different | 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... | MiniMax-M2 is a compact, high-efficiency large language model optimized for end-to-end coding and agentic workflows. With 10 billion activated parameters (230 billion total), it delivers near-frontier… | MiniMax-M3 is a multimodal foundation model from MiniMax. It supports text, image, and video inputs with text output, a 1M-token context window, and is suited for long-horizon agentic work, coding,... |
| Date | Feb 12, 2026Listed | Oct 23, 2025Listed | May 31, 2026Listed |
API prices USD · provider configuration shown below | |||
| Listed rate | $0.27 in / $0.95 out/ 1M tokensVenice · via OpenRouter · all routes ↗venice | $0.255 in / $1.02 out/ 1M tokensMinimax · via OpenRouter · all routes ↗minimax/fp8 · quantization=fp8 | $0.224 in / $0.88 out/ 1M tokensDeepInfra · via OpenRouter · all routes ↗deepinfra/fp8/flex · tier=flex · quantization=fp8 |
| Provider endpoints | 12 endpoints → | 3 endpoints → | 15 endpoints → |
Specifications | |||
| Context window | 204.8K | 204.8K | 1M |
| Max output | 32.8K | 131.1K | 512K |
| Parameters | — | — | — |
| Tool calling | Supported | Supported | Supported |
| Structured output | Supported | Supported | Supported |
| Image input | Not reported | Not reported | Supported |
| Video input | Not reported | Not reported | Supported |
| Reasoning | Supported | Supported | Supported |
| Prompt caching | Supported | Not reported | Supported |
| Open weights | Supported | Supported | Supported |