Google: Gemma 4 31B vs Qwen: Qwen3.5 397B A17B
Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-11-07.
| Google: Gemma 4 31B | Qwen: Qwen3.5 397B A17B | |
|---|---|---|
| Vendor | qwen | |
| Quality Score | 100 | 100 |
| Benchmark Score | 53.7 | 60.1 |
| Input Price | $0.20/M | $0.50/M |
| Output Price | $0.40/M | $3.60/M |
| Context Window | 262,144 | 262,144 |
| Max Output | 16,384 | 235,929 |
| Tool Calling | ✓ | ✓ |
| Structured Output | ✓ | ✓ |
| Reasoning Mode | ✓ | ✓ |
| Vision | ✓ | ✓ |
| Audio | - | - |
| Benchmark Scores | ||
| ai_index | 48.4 | 55.6 |
| ai_index_agentic | 23.8 | 32.7 |
| ai_index_coding | 71.7 | 79.5 |
| eqbench | 57.5 | 58.0 |
The verdict
For a sample job of 1 million input tokens and 250,000 output tokens, Gemma 4 31B costs $0.30 and Qwen3.5 397B A17B costs $1.40, so Gemma 4 31B is about 4.7 times cheaper for the same work.
Both read up to 262,144 tokens at once, roughly 393 pages of text.
On our blended benchmark score Qwen3.5 397B A17B leads 60.1 to 53.7.
Across the 20 tasks where they differ, Gemma 4 31B scores higher on 2 (Code Completion, Bulk Data Labeling), Qwen3.5 397B A17B on 18 (SQL Generation, Code Review, Code Refactoring and others).
Gemma 4 31B was released in April 2026 and Qwen3.5 397B A17B in February 2026.
Pick Gemma 4 31B if you need lower cost; pick Qwen3.5 397B A17B if you need benchmark performance or the stronger all-round task scores.
Costs use current list prices per million tokens; page counts assume about 0.75 words per token and 500 words per page.
Who wins by task?
| Task | Google: Gemma 4 31B | Qwen: Qwen3.5 397B A17B |
|---|---|---|
| SQL Generation | 156 | 159 |
| Code Review | 152 | 156 |
| Code Completion | 131 | 131 |
| Code Refactoring | 151 | 153 |
| Bug Fixing | 160 | 164 |
| Unit Test Generation | 143 | 145 |
| Code Documentation | 137 | 137 |
| Regex Writing | 131 | 132 |
| CI/CD Pipelines | 135 | 137 |
| Frontend Component Design | 138 | 140 |
| Data Analysis | 152 | 156 |
| CSV / Spreadsheet Cleanup | 145 | 146 |
| ETL Scripting | 143 | 145 |
| JSON Extraction | 143 | 144 |
| Bulk Data Labeling | 133 | 133 |
| OCR / Document Parsing | 140 | 141 |
| Table Extraction from PDFs | 140 | 141 |
| Long-Document Summarization | 149 | 150 |
| Short-Form Summarization | 129 | 129 |
| Blog Post Writing | 133 | 134 |
Scores reflect capability match + benchmark data + pricing for each task. Methodology →