Claude Sonnet 5.5 (max) vs Claude Haiku 5.5 (max)
Results updated 3 November 2026 · Both in Claude Code · How we benchmark
Verdict
Not comparable yet: Claude Sonnet 5.5 (max) is not yet ranked.
- Cost per task
- $0.54 vs $0.079
- Time per task
- 5.6 min (estimated) vs 7.8 min (estimated)
- Tokens per attempt
- 649k vs 859k
- Image input
- Yes vs Yes
Claude Sonnet 5.5 (max), Claude Code scores higher on GS-QA2 and Vision agents; Claude Haiku 5.5 (max), Claude Code scores higher on Earth-Bench V2 and GeoBenchX. Claude Haiku 5.5 (max), Claude Code is cheaper ($0.079 against $0.54 per task), but Claude Sonnet 5.5 (max), Claude Code is faster (5.6 min against 7.8 min).
Not yet ranked. Claude Sonnet 5.5 (max), Claude Code. Not in the Agent Index, cost and time charts until it has a score on every index evaluation. Results so far are in the side-by-side table.
Where they differ most
By kind of work, share of tasks passed, 0 to 100 · Higher is better
| Kind of work | Claude Sonnet 5.5 (max) | Claude Haiku 5.5 (max) |
|---|---|---|
| Spatial statistics | 100 | 45 |
| Climate and time series | 58 | 76 |
| Recognising infeasible requests | 75 | 86 |
| Remote sensing and raster analysis | 88 | 77 |
Where the tasks are set: each kind of work links to its tasks and scoresCharts for each: Claude Sonnet 5.5 (max), Claude Code · Claude Haiku 5.5 (max), Claude Code.
Side by side
| Metric | Claude Sonnet 5.5 (max), Claude Code | Claude Haiku 5.5 (max), Claude Code |
|---|---|---|
| Geospatial Agent Index: Text only | 75 | 78 |
| Geospatial Agent Index: Text + vision | No data | 79 |
| Earth-Bench V2 score | 57 | 70 |
| GeoBenchX score | 77 | 78 |
| GS-QA2 score | 92 | 87 |
| Vision agents score | 89 | 81 |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 100 | 97 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 100 | 100 |
| OGD4All score (standalone; in no index) | 97 | 97 |
| GeoNatureAgent score (standalone; in no index) | 100 | 100 |
| MultiGlobeQA score (standalone; in no index) | 78 | 76 |
| GeoAgentEval score (standalone; in no index) | 100 | 96 |
| Remote sensing and raster analysis score | 88 | 77 |
| GIS analysis score | 95 | 91 |
| Climate and time series score | 58 | 76 |
| Spatial statistics score | 100 | 45 |
| Making maps score | 78 | 85 |
| Recognising infeasible requests score | 75 | 86 |
| Cost per task | $0.54 | $0.079 |
| Time per task (at 100 output tokens/s) | 5.6 min (estimated) | 7.8 min (estimated) |
| Tokens per attempt | 649k | 859k |
| Turns per attempt | 12.8 | 13.1 |
| Input price per 1M tokens | $2.00 | $0.10 |
| Output price per 1M tokens | $10.00 | $0.50 |
| Context window | 1,000,000 | 1,000,000 |
| Image input | Yes | Yes |
| Developer | Anthropic | Anthropic |