Mistral Small 4 vs Qwen3.6 Flash
Price and specs side by side · USD per 1M tokens · verified 2026-10-11
// side by side
| metric | Mistral Small 4 | Qwen3.6 Flash |
|---|
| input / 1M | $0.15 | $0.1875 |
| output / 1M | $0.60 | $1.125 |
| blended (3:1) | $0.2625 | $0.4219 |
| cache read / 1M | $0.015 | — |
| context | 262.1K | 1M |
| max output | 256K | 65.5K |
| input types | text · image | text · image · video |
| reasoning | yes | yes |
| tool calling | yes | yes |
| structured output | no | yes |
| open weights | yes | no |
| released | 2026-03-16 | 2026-04-27 |
| knowledge cutoff | 2025-06 | n/a |
| lab | Mistral | Alibaba (Qwen) |
// monthly cost scenarios
| workload | tokens in / out | Mistral Small 4 | Qwen3.6 Flash |
|---|
| Chatbot | 10M / 2M | $2.70 | $4.13 |
| RAG app | 50M / 5M | $10.50 | $15.00 |
| Coding agent | 20M / 10M | $9.00 | $15.00 |
// verdict
- Mistral Small 4 is 37.8% cheaper than Qwen3.6 Flash on a blended 3:1 input/output basis.
- Qwen3.6 Flash has the larger context window (1M vs 262.1K tokens).
- Only Qwen3.6 Flash supports structured output.