Tools
Compare models
Add up to four models via ?models=slug-a,slug-b — or pick from any model page.
| Spec | Moonshot AI | OpenAI |
|---|---|---|
| Params | 1.1T / 32B | 120B |
| Architecture | moe | moe |
| Context | 262K | 131K |
| Weights | ||
| License | Custom licence | Apache License 2.0 |
| 667.5 GBest | 75.9 GBest | |
| Cheapest hosted (out/1M) | $2.48 · OpenRouter | $0.17 · OpenRouter |
| Released | Apr 14, 2026 | Aug 4, 2025 |
| LiveBench | 70.5via Thinking | — |
| LiveBench Agentic Coding | 46.9via Thinking | — |
| LiveBench Coding | 78.6via Thinking | — |
| LiveBench Data Analysis | 65.1via Thinking | — |
| LiveBench Instruction Following | 64.4via Thinking | — |
| LiveBench Language | 75.1via Thinking | — |
| LiveBench Mathematics | 84.3via Thinking | — |
| LiveBench Reasoning | 79.4via Thinking | — |
| Arena Agent (IPS) | −0.011 | — |
| Arena Coding | 1514.8 | 1390.1 |
| Arena Creative Writing | 1429.5 | 1277.5 |
| Arena Hard Prompts | 1484.8 | 1362 |
| Arena Instruction Following | 1454.2 | 1325.3 |
| Arena Maths | 1479.3 | 1381.3 |
| Arena Text (overall) | 1460.7 | 1352.3 |
| Arena Code (WebDev) | 1509.5 | — |
| SWE-bench Verified | — | 26via mini-SWE-agent |
