Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | DeepSeek | Moonshot AI |
|---|---|---|
| Params | 685B / 37B | 1.1T / 32B |
| Architecture | moe | moe |
| Context | 164K | 262K |
| Weights | ||
| License | MIT License | Custom licence |
| 432.1 GBest | 667.5 GBest | |
| Cheapest hosted (out/1M) | $0.40 · Novita AI | $3.50 · OpenRouter |
| Released | Dec 1, 2025 | Jun 11, 2026 |
| Arena Coding | 1469.6 | — |
| Arena Creative Writing | 1400.4 | — |
| Arena Hard Prompts | 1447.2 | — |
| Arena Instruction Following | 1420.6 | — |
| Arena Maths | 1428.5 | — |
| Arena Text (overall) | 1425 | — |
| Arena Code (WebDev) | 1323.4 | 1472.6 |
| SWE-bench Verified | 70via mini-SWE-agent | — |
| LiveBench | — | 68.4 |
| LiveBench Agentic Coding | — | 45.7 |
| LiveBench Coding | — | 74 |
| LiveBench Data Analysis | — | 62.7 |
| LiveBench Instruction Following | — | 56.3 |
| LiveBench Language | — | 77.9 |
| LiveBench Mathematics | — | 79.6 |
| LiveBench Reasoning | — | 82.8 |
| Arena Agent (IPS) | — | 0.004 |
