Qwen3.5 122B-A10B vs Qwen2.5 72B
Qwen2.5 72B is cheaper on output tokens, while Qwen3.5 122B-A10B offers a larger context window. Choose Qwen3.5 122B-A10B or Qwen2.5 72B based on the trade-off between cost, context, and the benchmarks that matter for your use case.
| Spec | Qwen3.5 122B-A10B | Qwen2.5 72B |
|---|---|---|
| Provider | Alibaba | NanoGPT |
| Input / 1M tokens | $0.40 | $0.36 |
| Output / 1M tokens | $3.20 | $0.40 |
| Context window | 262K | 131K |
| Parameters | 122B | 73B |
| Open weights | Yes | Yes |
| Released | Feb 2026 | Jul 2025 |
| LMArena (Chatbot Arena) Elo score | 1417 | — |
| LMArena WebDev Elo score | 1361 | — |
| LMArena Vision Elo score | 1245 | — |
| Epoch Capabilities Index score | — | 129 |
FAQ
Is Qwen3.5 122B-A10B or Qwen2.5 72B cheaper?
Qwen2.5 72B is cheaper on output tokens ($0.40 vs $3.20 per 1M).
Which has the larger context window, Qwen3.5 122B-A10B or Qwen2.5 72B?
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.