Qwen3.5 122B-A10B vs MiniMax 2.5
Qwen3.5 122B-A10B is cheaper on output tokens, while Qwen3.5 122B-A10B offers a larger context window. Choose Qwen3.5 122B-A10B or MiniMax 2.5 based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Qwen3.5 122B-A10B | MiniMax 2.5 |
|---|---|---|
| Provider | Alibaba | Regolo AI |
| Input / 1M tokens | $0.40 | $0.80 |
| Output / 1M tokens | $3.20 | $3.50 |
| Context window | 262K | 190K |
| Parameters | 122B | — |
| Open weights | Yes | No |
| Released | Feb 2026 | Mar 2026 |
| LMArena (Chatbot Arena) Elo score | 1417 | — |
| LMArena WebDev Elo score | 1361 | — |
| LMArena Vision Elo score | 1245 | — |
| SWE-bench Multilingual score | — | 68.3 |
FAQ
Is Qwen3.5 122B-A10B or MiniMax 2.5 cheaper?
Qwen3.5 122B-A10B is cheaper on output tokens ($3.20 vs $3.50 per 1M).
Which has the larger context window, Qwen3.5 122B-A10B or MiniMax 2.5?
Qwen3.5 122B-A10B has the larger context window (262K tokens).
Pricing is indicative — confirm with the provider before production use. Highlighted values indicate the better figure for that row.