Anthropic: Claude Sonnet 5 vs Google: Gemma 4 31B (free)
Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-11-07.
| Anthropic: Claude Sonnet 5 | Google: Gemma 4 31B (free) | |
|---|---|---|
| Vendor | anthropic | |
| Quality Score | 100 | 100 |
| Benchmark Score | 91.4 | 53.7 |
| Input Price | $2.00/M | Free |
| Output Price | $10.00/M | Free |
| Context Window | 1,000,000 | 262,144 |
| Max Output | 128,000 | 32,768 |
| Tool Calling | ✓ | ✓ |
| Structured Output | ✓ | ✓ |
| Reasoning Mode | ✓ | ✓ |
| Vision | ✓ | ✓ |
| Audio | - | - |
| Benchmark Scores | ||
| ai_index | 88.0 | 48.4 |
| ai_index_agentic | 77.0 | 23.8 |
| ai_index_coding | 100.0 | 71.7 |
| eqbench | - | 57.5 |
The verdict
For a sample job of 1 million input tokens and 250,000 output tokens, Gemma 4 31B (free) costs $0 and Claude Sonnet 5 costs $4.50, so Gemma 4 31B (free) is free for the same work.
Claude Sonnet 5 reads up to 1,000,000 tokens at once, roughly 1,500 pages of text, against about 393 pages for Gemma 4 31B (free).
Only Gemma 4 31B (free) offers video input.
On our blended benchmark score Claude Sonnet 5 leads 91.4 to 53.7.
Across the 20 tasks where they differ, Claude Sonnet 5 scores higher on 16 (SQL Generation, Code Review, Code Refactoring and others), Gemma 4 31B (free) on 4 (Code Completion, JSON Extraction, Bulk Data Labeling and others).
Claude Sonnet 5 was released in June 2026 and Gemma 4 31B (free) in April 2026.
Pick Claude Sonnet 5 if you need long documents, benchmark performance or the stronger all-round task scores; pick Gemma 4 31B (free) if you need lower cost or video input.
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 | Anthropic: Claude Sonnet 5 | Google: Gemma 4 31B (free) |
|---|---|---|
| SQL Generation | 171 | 156 |
| Code Review | 167 | 152 |
| Code Completion | 119 | 131 |
| Code Refactoring | 164 | 151 |
| Bug Fixing | 180 | 160 |
| Unit Test Generation | 154 | 143 |
| Code Documentation | 141 | 137 |
| Regex Writing | 135 | 131 |
| CI/CD Pipelines | 144 | 135 |
| Frontend Component Design | 146 | 138 |
| Data Analysis | 171 | 152 |
| CSV / Spreadsheet Cleanup | 153 | 145 |
| ETL Scripting | 154 | 143 |
| JSON Extraction | 138 | 143 |
| Bulk Data Labeling | 124 | 133 |
| OCR / Document Parsing | 146 | 140 |
| Table Extraction from PDFs | 146 | 140 |
| Long-Document Summarization | 159 | 149 |
| Short-Form Summarization | 122 | 129 |
| Blog Post Writing | 140 | 133 |
Scores reflect capability match + benchmark data + pricing for each task. Methodology →