Axis Spatial

Claude Code – Claude Opus 5.5 (high) vs. Codex – GPT-6 Luna (xhigh): geospatial agent comparison

Comparison between Claude Code – Claude Opus 5.5 (high) and Codex – GPT-6 Luna (xhigh) on the Geospatial Agent Index, each benchmark, cost, time and token use, from the same tasks, agent and limits. Methodology.

Side by side

Claude Code – Claude Opus 5.5 (high) vs. Codex – GPT-6 Luna (xhigh)
MetricClaude Code – Claude Opus 5.5 (high)Codex – GPT-6 Luna (xhigh)
Geospatial Agent Index83* (partial: 1.7 of 3 attempts)77* (partial: 1.9 of 3 attempts)
GeoAgentBench score9699
GeoBenchX score6660
Earth-Bench score6955
GeoAnalystBench score10095
Raster and remote sensing score8475
Vector and overlay score8378
Networks and routing score100100
Climate and time series score7975
Spatial statistics and interpolation score6263
Mapping and cartography score7664
Recognising infeasible requests score5962
Cost per task$0.175$0.0092
Time per task1.0 min3.4 min
Tokens per attempt127k236k
Turns per attempt5.79.8
Attempts decided422512
Input price per 1M tokens$4.00$0.10
Output price per 1M tokens$20.00$0.50
Context windowNo dataNo data
Image inputNo dataNo data
DeveloperAnthropicOpenAI

Charts

Geospatial Agent Index

Equal-weight average of the evaluation scores, 0 to 100 · Higher is better

  • Partial coverage
Geospatial Agent IndexClaude Code – Claude Opus 5.5 (high): 83* (partial coverage); Codex – GPT-6 Luna (xhigh): 77* (partial coverage)025507510083*Claude Opus 5.5 (high)77*GPT-6 Luna (xhigh)
Data table: Geospatial Agent Index
Geospatial Agent Index
ModelCreatorGeospatial Agent IndexRange while attempts are pendingBenchmarks coveredCoverage (attempts)
Claude Code – Claude Opus 5.5 (high)Anthropic83* (partial: 1.7 of 3 attempts)71 to 834 of 4422 of 870 planned attempts
Codex – GPT-6 Luna (xhigh)OpenAI77* (partial: 1.9 of 3 attempts)72 to 784 of 4512 of 870 planned attempts

Cost per task

Average cost per task (USD) · Lower is better

  • Partial coverage
Cost per taskCodex – GPT-6 Luna (xhigh): $0.0092 (partial coverage); Claude Code – Claude Opus 5.5 (high): $0.175 (partial coverage)$0$0.050$0.100$0.150$0.200$0.0092GPT-6 Luna (xhigh)$0.175Claude Opus 5.5 (high)
Data table: Cost per task
Cost per task
ModelCreatorCost per taskCoverage (attempts)
Claude Code – Claude Opus 5.5 (high)Anthropic$0.175496 of 870 planned attempts
Codex – GPT-6 Luna (xhigh)OpenAI$0.0092560 of 870 planned attempts

Time per task

Average agent wall time per task · Lower is better

  • Partial coverage
Time per taskClaude Code – Claude Opus 5.5 (high): 1.0 min (partial coverage); Codex – GPT-6 Luna (xhigh): 3.4 min (partial coverage)0m1m2m3m4m1.0 minClaude Opus 5.5 (high)3.4 minGPT-6 Luna (xhigh)
Data table: Time per task
Time per task
ModelCreatorTime per taskCoverage (attempts)
Claude Code – Claude Opus 5.5 (high)Anthropic1.0 min496 of 870 planned attempts
Codex – GPT-6 Luna (xhigh)OpenAI3.4 min560 of 870 planned attempts

Frequently asked questions

Which scores higher on the Geospatial Agent Index, Claude Code – Claude Opus 5.5 (high) or Codex – GPT-6 Luna (xhigh)?
Claude Code – Claude Opus 5.5 (high), at 83 against 77 for Codex – GPT-6 Luna (xhigh). Claude Code – Claude Opus 5.5 (high): 422 of 870 planned attempts. Codex – GPT-6 Luna (xhigh): 512 of 870 planned attempts. Coverage is partial; scores can change as more attempts are decided.
Which is cheaper per task, Claude Code – Claude Opus 5.5 (high) or Codex – GPT-6 Luna (xhigh)?
Codex – GPT-6 Luna (xhigh), at $0.0092 against $0.175 for Claude Code – Claude Opus 5.5 (high).
Which is faster per task, Claude Code – Claude Opus 5.5 (high) or Codex – GPT-6 Luna (xhigh)?
Claude Code – Claude Opus 5.5 (high), at 1.0 min against 3.4 min for Codex – GPT-6 Luna (xhigh).
Which has the larger context window, Claude Code – Claude Opus 5.5 (high) or Codex – GPT-6 Luna (xhigh)?
Not known yet: Claude Code – Claude Opus 5.5 (high) has no context window data and Codex – GPT-6 Luna (xhigh) has no context window data.

More: Claude Code – Claude Opus 5.5 (high) · Codex – GPT-6 Luna (xhigh) · all comparisons