head-to-head
Mistral: Mistral Small 4 vs Qwen: Qwen3.5-27B
Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-09-23.
| Mistral: Mistral Small 4 | Qwen: Qwen3.5-27B | |
|---|---|---|
| Vendor | mistralai | qwen |
| Quality Score | 100 | 100 |
| Benchmark Score | 3.8 | 35.5 |
| Input Price | $0.15/M | $0.20/M |
| Output Price | $0.60/M | $1.56/M |
| Context Window | 262,144 | 262,144 |
| Max Output | 209,715 | 65,536 |
| Tool Calling | ✓ | ✓ |
| Structured Output | ✓ | ✓ |
| Reasoning Mode | ✓ | ✓ |
| Vision | ✓ | ✓ |
| Audio | - | - |
| Benchmark Scores | ||
| ai_index | 9.6 | 37.8 |
Who wins by task?
| Task | Mistral: Mistral Small 4 | Qwen: Qwen3.5-27B |
|---|---|---|
| SQL Generation | 132 | 135 |
| Code Review | 128 | 133 |
| Code Completion | 129 | 129 |
| Code Refactoring | 129 | 133 |
| Bug Fixing | 132 | 137 |
| Unit Test Generation | 122 | 125 |
| Code Documentation | 127 | 129 |
| Regex Writing | 120 | 123 |
| CI/CD Pipelines | 118 | 121 |
| Frontend Component Design | 123 | 126 |
| Data Analysis | 125 | 129 |
| CSV / Spreadsheet Cleanup | 128 | 129 |
| ETL Scripting | 123 | 127 |
| JSON Extraction | 131 | 131 |
| Bulk Data Labeling | 129 | 129 |
| OCR / Document Parsing | 128 | 130 |
| Table Extraction from PDFs | 128 | 130 |
| Long-Document Summarization | 131 | 135 |
| Short-Form Summarization | 124 | 125 |
| Blog Post Writing | 120 | 123 |
Scores reflect capability match + benchmark data + pricing for each task. Methodology →
Related comparisons
Anthropic: Claude Haiku Latest vs Mistral: Mistral Small 4
Anthropic: Claude Haiku Latest vs Qwen: Qwen3.5-27B
Anthropic: Claude Sonnet 5 vs Mistral: Mistral Small 4
Anthropic: Claude Sonnet 5 vs Qwen: Qwen3.5-27B
Anthropic: Claude Sonnet Latest vs Mistral: Mistral Small 4
Anthropic: Claude Sonnet Latest vs Qwen: Qwen3.5-27B
ByteDance Seed: Seed-2.0-Lite vs Qwen: Qwen3.5-27B
ByteDance Seed: Seed-2.0-Mini vs Qwen: Qwen3.5-27B