Axis Spatial

Terminus-2 – Gemini 3.1 Pro (preview) (high) vs. Codex – GPT-6.1 Sol (medium): geospatial agent comparison

Comparison between Terminus-2 – Gemini 3.1 Pro (preview) (high) and Codex – GPT-6.1 Sol (medium) on the Geospatial Agent Index, each benchmark, cost, time and token use, from the same tasks, agent and limits. Methodology.

Side by side

Terminus-2 – Gemini 3.1 Pro (preview) (high) vs. Codex – GPT-6.1 Sol (medium)
MetricTerminus-2 – Gemini 3.1 Pro (preview) (high)Codex – GPT-6.1 Sol (medium)
Geospatial Agent Index75 (attempts per task: 3)79* (attempts per task: 2.6)
GeoAgentBench score94100
GeoBenchX score6567
Earth-Bench score5451
GeoAnalystBench score86100
Raster and remote sensing score6771
Vector and overlay score7882
Networks and routing score93100
Climate and time series score8275
Spatial statistics and interpolation score7262
Mapping and cartography score5775
Recognising infeasible requests score7569
Cost per task$0.252$0.060
Time per task3.4 min1.4 min
Tokens per attempt73k111k
Turns per attempt7.16.6
Attempts decided870749
Input price per 1M tokens$2.00$2.00
Output price per 1M tokens$12.00$10.00
Context window1,048,576No data
Image inputYesNo data
DeveloperGoogleOpenAI

Charts

Geospatial Agent Index

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

  • Partial coverage
Geospatial Agent IndexCodex – GPT-6.1 Sol (medium): 79* (partial coverage); Terminus-2 – Gemini 3.1 Pro (preview) (high): 75025507510079*GPT-6.1 Sol (medium)75Gemini 3.1 Pro (preview) (high)
Data table: Geospatial Agent Index
Geospatial Agent Index
ModelCreatorGeospatial Agent IndexRange while attempts are pendingBenchmarks coveredCoverage (attempts)
Codex – GPT-6.1 Sol (medium)OpenAI79* (attempts per task: 2.6)No pending judgements; coverage incomplete4 of 4749 of 870 planned attempts (6 excluded)
Terminus-2 – Gemini 3.1 Pro (preview) (high)Google75 (attempts per task: 3)No pending judgements4 of 4870 of 870 planned attempts

Cost per task

Average cost per task (USD) · Lower is better

  • Partial coverage
Cost per taskCodex – GPT-6.1 Sol (medium): $0.060 (partial coverage); Terminus-2 – Gemini 3.1 Pro (preview) (high): $0.252$0$0.075$0.150$0.225$0.300$0.060GPT-6.1 Sol (medium)$0.252Gemini 3.1 Pro (preview) (high)
Data table: Cost per task
Cost per task
ModelCreatorCost per taskCoverage (attempts)
Terminus-2 – Gemini 3.1 Pro (preview) (high)Google$0.252870 of 870 planned attempts
Codex – GPT-6.1 Sol (medium)OpenAI$0.060749 of 870 planned attempts (6 excluded)

Time per task

Average agent wall time per task · Lower is better

  • Partial coverage
Time per taskCodex – GPT-6.1 Sol (medium): 1.4 min (partial coverage); Terminus-2 – Gemini 3.1 Pro (preview) (high): 3.4 min0m1m2m3m4m1.4 minGPT-6.1 Sol (medium)3.4 minGemini 3.1 Pro (preview) (high)
Data table: Time per task
Time per task
ModelCreatorTime per taskCoverage (attempts)
Terminus-2 – Gemini 3.1 Pro (preview) (high)Google3.4 min870 of 870 planned attempts
Codex – GPT-6.1 Sol (medium)OpenAI1.4 min749 of 870 planned attempts (6 excluded)

Frequently asked questions

Which scores higher on the Geospatial Agent Index, Terminus-2 – Gemini 3.1 Pro (preview) (high) or Codex – GPT-6.1 Sol (medium)?
Codex – GPT-6.1 Sol (medium), at 79 against 75 for Terminus-2 – Gemini 3.1 Pro (preview) (high). Terminus-2 – Gemini 3.1 Pro (preview) (high): 870 of 870 planned attempts. Codex – GPT-6.1 Sol (medium): 749 of 870 planned attempts (6 excluded). Coverage is partial; scores can change as more attempts are decided.
Which is cheaper per task, Terminus-2 – Gemini 3.1 Pro (preview) (high) or Codex – GPT-6.1 Sol (medium)?
Codex – GPT-6.1 Sol (medium), at $0.060 against $0.252 for Terminus-2 – Gemini 3.1 Pro (preview) (high).
Which is faster per task, Terminus-2 – Gemini 3.1 Pro (preview) (high) or Codex – GPT-6.1 Sol (medium)?
Codex – GPT-6.1 Sol (medium), at 1.4 min against 3.4 min for Terminus-2 – Gemini 3.1 Pro (preview) (high).
Which has the larger context window, Terminus-2 – Gemini 3.1 Pro (preview) (high) or Codex – GPT-6.1 Sol (medium)?
Not known yet: Codex – GPT-6.1 Sol (medium) has no context window data.

More: Terminus-2 – Gemini 3.1 Pro (preview) (high) · Codex – GPT-6.1 Sol (medium) · all comparisons