MiniMax: MiniMax M2.5
Chatminimax/minimax-m2.5MiniMax M2.5 is MiniMax's state-of-the-art large language model designed for real-world productivity. Trained across diverse, complex digital working environments, it builds on M2.1's coding expertise and extends into general office work: generating and operating Word, Excel, and PowerPoint files, switching context between different software environments, and working across agent and human teams. It scores 80.2% on SWE-Bench Verified, 51.3% on Multi-SWE-Bench, and 76.3% on BrowseComp, and is more token-efficient than previous generations because it was trained to plan its actions and output. Reasoning, function calling, prompt caching, and web search are supported. Context: 200K tokens, output: 131K. Available via OpenAI and Anthropic protocols.
Providers
/v1/chat/completionsCode Examples
from openai import OpenAIclient = OpenAI(base_url="https://api.ofox.run/v1",api_key="YOUR_OFOX_API_KEY",)response = client.chat.completions.create(model="minimax/minimax-m2.5",messages=[{"role": "user", "content": "Hello!"}],)print(response.choices[0].message.content)
Benchmarks
MiniMax: MiniMax M2.5 scores 1391 in the Overall category of the LMArena text leaderboard (style control), ranking #137 of 374 models based on 41,159 human preference votes (updated 2026-07-12).
| Category | Arena Score | 95% CI | Rank | Votes |
|---|---|---|---|---|
| Overall | 1391 | 1387–1395 | #137 of 374 | 41,159 |
| Hard Prompts | 1416 | 1411–1420 | #133 of 374 | 25,585 |
| Coding | 1444 | 1438–1451 | #126 of 369 | 10,895 |
| Math | 1396 | 1384–1408 | #132 of 362 | 2,436 |
| Creative Writing | 1359 | 1350–1367 | #135 of 372 | 6,411 |
| Instruction Following | 1381 | 1375–1387 | #140 of 374 | 12,998 |
| Chinese | 1429 | 1416–1443 | #125 of 344 | 2,187 |
Source: LMArena · CC BY 4.0 · Updated 2026-07-12 · Methodology ↗ · Ranks compare models within each category of the LMArena text leaderboard (style control). Scores come from third-party human preference evaluations, not from OFOX.