Anthropic: Claude Haiku Latest vs Google: Gemma 4 31B
Side-by-side comparison of specs, pricing, benchmark scores, and task rankings. Updated 2026-11-07.
| Anthropic: Claude Haiku Latest | Google: Gemma 4 31B | |
|---|---|---|
| Vendor | ~anthropic | |
| Quality Score | 100 | 100 |
| Benchmark Score | - | 53.7 |
| Input Price | $0.10/M | $0.20/M |
| Output Price | $0.50/M | $0.40/M |
| Context Window | 1,000,000 | 262,144 |
| Max Output | 128,000 | 16,384 |
| Tool Calling | ✓ | ✓ |
| Structured Output | ✓ | ✓ |
| Reasoning Mode | ✓ | ✓ |
| Vision | ✓ | ✓ |
| Audio | - | - |
| Benchmark Scores | ||
| ai_index | - | 48.4 |
| ai_index_agentic | - | 23.8 |
| ai_index_coding | - | 71.7 |
| eqbench | - | 57.5 |
The verdict
For a sample job of 1 million input tokens and 250,000 output tokens, Claude Haiku Latest costs $0.23 and Gemma 4 31B costs $0.30, so Claude Haiku Latest is about 1.3 times cheaper for the same work.
Claude Haiku Latest reads up to 1,000,000 tokens at once, roughly 1,500 pages of text, against about 393 pages for Gemma 4 31B.
Only Gemma 4 31B offers video input.
Across the 20 tasks where they differ, Claude Haiku Latest scores higher on 1 (Code Completion), Gemma 4 31B on 19 (SQL Generation, Code Review, Code Refactoring and others).
Pick Claude Haiku Latest if you need lower cost or long documents; pick Gemma 4 31B if you need video input 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 | Anthropic: Claude Haiku Latest | Google: Gemma 4 31B |
|---|---|---|
| SQL Generation | 133 | 156 |
| Code Review | 132 | 152 |
| Code Completion | 131 | 131 |
| Code Refactoring | 136 | 151 |
| Bug Fixing | 136 | 160 |
| Unit Test Generation | 124 | 143 |
| Code Documentation | 131 | 137 |
| Regex Writing | 119 | 131 |
| CI/CD Pipelines | 120 | 135 |
| Frontend Component Design | 122 | 138 |
| Data Analysis | 124 | 152 |
| CSV / Spreadsheet Cleanup | 133 | 145 |
| ETL Scripting | 128 | 143 |
| JSON Extraction | 131 | 143 |
| Bulk Data Labeling | 129 | 133 |
| OCR / Document Parsing | 131 | 140 |
| Table Extraction from PDFs | 131 | 140 |
| Long-Document Summarization | 137 | 149 |
| Short-Form Summarization | 123 | 129 |
| Blog Post Writing | 121 | 133 |
Scores reflect capability match + benchmark data + pricing for each task. Methodology →