Codex – GPT-5.6 Luna (high) vs. Codex – GPT-6 Luna (low): geospatial agent comparison

Comparison between Codex – GPT-5.6 Luna (high) and Codex – GPT-6 Luna (low) on the Geospatial Agent Index, each benchmark, cost, time and token use, from the same tasks, agent and limits. Methodology.

Side by side

Codex – GPT-5.6 Luna (high) vs. Codex – GPT-6 Luna (low)
Metric	Codex – GPT-5.6 Luna (high)	Codex – GPT-6 Luna (low)
Geospatial Agent Index	72* (partial: attempt 1 on 469 of 488 tasks)	51* (partial: attempt 1 on 449 of 488 tasks)
Earth-Bench V2 score	67	44
GeoBenchX score	68	50
GS-QA2 score	81	59
GeoAgentBench score (guided GIS workflow; outside the index)	98	92
GeoAnalystBench score (guided GIS workflow; outside the index)	100	79
Raster and remote sensing score	78	55
Vector and overlay score	79	57
Climate and time series score	69	49
Spatial statistics and interpolation score	40	38
Mapping and cartography score	87	71
Recognising infeasible requests score	53	30
Cost per task	$0.016	$0.0031
Time per task	3.1 min	1.4 min
Tokens per attempt	251k	114k
Turns per attempt	12.2	7.8
Attempts decided	469	449
Input price per 1M tokens	$0.20	$0.10
Output price per 1M tokens	$1.20	$0.50
Context window	No data	No data
Image input	No data	No data
Developer	OpenAI	OpenAI

Charts

Partial coverage. Codex – GPT-5.6 Luna (high): 131 of 150 GeoBenchX tasks; Codex – GPT-6 Luna (low): 111 of 150 GeoBenchX tasks. Grey placeholders identify these entries in the index, cost and output token charts; their height does not represent a value. Comparable bars appear when first-attempt coverage is complete. Their available results remain in the side-by-side table.

Geospatial Agent Index

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

Data table: Geospatial Agent Index

Geospatial Agent Index
Model	Creator	Geospatial Agent Index	Range while attempts are pending	Benchmarks covered	Coverage (attempts)

Cost per task

Average cost per task (USD) · Lower is better

How to read this chart

Not shown: partial coverage. 2 entries without a first attempt on every index task. What is still running is on the models page.

Data table: Cost per task

Cost per task
Model	Creator	Cost per task	Coverage (attempts)

Output tokens per task

Average output tokens per attempt, reasoning included · Lower is better

How to read this chart

Not shown: partial coverage. 2 entries without a first attempt on every index task. What is still running is on the models page.

Data table: Output tokens per task

Output tokens per task
Model	Creator	Output tokens per task	Coverage (attempts)

Frequently asked questions

Which scores higher on the Geospatial Agent Index, Codex – GPT-5.6 Luna (high) or Codex – GPT-6 Luna (low)?

Not comparable yet: Codex – GPT-5.6 Luna (high) has partial coverage (131 of 150 GeoBenchX tasks) and Codex – GPT-6 Luna (low) has partial coverage (111 of 150 GeoBenchX tasks), so it is not ranked. Codex – GPT-5.6 Luna (high): 469 of 722 scheduled attempts. Codex – GPT-6 Luna (low): 449 of 722 scheduled attempts. Coverage is partial; scores can change as more attempts are decided.

Which is cheaper per task, Codex – GPT-5.6 Luna (high) or Codex – GPT-6 Luna (low)?

Codex – GPT-6 Luna (low), at $0.0031 against $0.016 for Codex – GPT-5.6 Luna (high).

Which writes fewer output tokens per task, Codex – GPT-5.6 Luna (high) or Codex – GPT-6 Luna (low)?

Codex – GPT-6 Luna (low), at 1k against 5k for Codex – GPT-5.6 Luna (high).

Which has the larger context window, Codex – GPT-5.6 Luna (high) or Codex – GPT-6 Luna (low)?

Not known yet: Codex – GPT-5.6 Luna (high) has no context window data and Codex – GPT-6 Luna (low) has no context window data.

More: Codex – GPT-5.6 Luna (high) · Codex – GPT-6 Luna (low) · all comparisons