Claude Opus 5.5 (max) vs GPT-5.6 Sol (high)
Results updated 3 November 2026 · In Claude Code and Codex, so part of any gap can come from the harness · How we benchmark
Verdict
Not comparable yet: Claude Opus 5.5 (max) is not yet ranked.
- Cost per task
- $0.83 vs $0.21
- Time per task
- 4.5 min (estimated) vs 1.0 min
- Tokens per attempt
- 530k vs 180k
- Image input
- Yes vs Yes
Claude Opus 5.5 (max), Claude Code scores higher than GPT-5.6 Sol (high), Codex on all 4 evaluations scored for both. GPT-5.6 Sol (high), Codex is cheaper and faster: $0.21 against $0.83 per task, and 1.0 min against 4.5 min.
Not yet ranked. Claude Opus 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 Opus 5.5 (max) | GPT-5.6 Sol (high) |
|---|---|---|
| Recognising infeasible requests | 79 | 58 |
| Climate and time series | 60 | 76 |
| Spatial statistics | 50 | 62 |
| Remote sensing and raster analysis | 88 | 78 |
Where the tasks are set: each kind of work links to its tasks and scoresCharts for each: Claude Opus 5.5 (max), Claude Code · GPT-5.6 Sol (high), Codex.
Side by side
| Metric | Claude Opus 5.5 (max), Claude Code | GPT-5.6 Sol (high), Codex |
|---|---|---|
| Geospatial Agent Index: Text only | 79 | 75 |
| Geospatial Agent Index: Text + vision | No data | 75 |
| Earth-Bench V2 score | 71 | 68 |
| GeoBenchX score | 74 | 70 |
| GS-QA2 score | 92 | 87 |
| Vision agents score | 98 | 75 |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 100 | 100 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 100 | 89 |
| OGD4All score (standalone; in no index) | 95 | No data |
| GeoNatureAgent score (standalone; in no index) | 100 | No data |
| MultiGlobeQA score (standalone; in no index) | 76 | No data |
| GeoAgentEval score (standalone; in no index) | 100 | No data |
| Remote sensing and raster analysis score | 88 | 78 |
| GIS analysis score | 93 | 87 |
| Climate and time series score | 60 | 76 |
| Spatial statistics score | 50 | 62 |
| Making maps score | 77 | 82 |
| Recognising infeasible requests score | 79 | 58 |
| Cost per task | $0.83 | $0.21 |
| Time per task (at 100 output tokens/s) | 4.5 min (estimated) | 1.0 min |
| Tokens per attempt | 530k | 180k |
| Turns per attempt | 12.0 | 9.5 |
| Input price per 1M tokens | $4.00 | $4.00 |
| Output price per 1M tokens | $20.00 | $20.00 |
| Context window | 1,000,000 | 1,050,000 |
| Image input | Yes | Yes |
| Developer | Anthropic | OpenAI |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, Claude Opus 5.5 (max), Claude Code or GPT-5.6 Sol (high), Codex?
Not comparable yet: Claude Opus 5.5 (max), Claude Code is not yet ranked.
Which is cheaper per task, Claude Opus 5.5 (max), Claude Code or GPT-5.6 Sol (high), Codex?
GPT-5.6 Sol (high), Codex, $0.21 to $0.83.
Which writes fewer output tokens per task, Claude Opus 5.5 (max), Claude Code or GPT-5.6 Sol (high), Codex?
GPT-5.6 Sol (high), Codex, 4k to 23k.
Which has the larger context window, Claude Opus 5.5 (max), Claude Code or GPT-5.6 Sol (high), Codex?
GPT-5.6 Sol (high), Codex, 1,050,000 tokens to 1,000,000 tokens.