Terminus-2 – Kimi K2.6 (always on) vs. Codex – GPT-6 Sol (high): geospatial agent comparison

Comparison between Terminus-2 – Kimi K2.6 (always on) and Codex – GPT-6 Sol (high) 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 – Kimi K2.6 (always on) vs. Codex – GPT-6 Sol (high)
Metric	Terminus-2 – Kimi K2.6 (always on)	Codex – GPT-6 Sol (high)
Geospatial Agent Index	70* (partial: attempt 1 on 485 of 488 tasks)	80* (partial: attempt 1 on 475 of 488 tasks)
Earth-Bench V2 score	68	76
GeoBenchX score	64	75
GS-QA2 score	77	89
GeoAgentBench score (guided GIS workflow; outside the index)	90	100
GeoAnalystBench score (guided GIS workflow; outside the index)	79	100
Raster and remote sensing score	77	84
Vector and overlay score	74	86
Climate and time series score	69	80
Spatial statistics and interpolation score	15	33
Mapping and cartography score	77	85
Recognising infeasible requests score	57	79
Cost per task	$0.089	$0.080
Time per task	8.0 min	1.7 min
Tokens per attempt	173k	144k
Turns per attempt	12.7	8.6
Attempts decided	485	460
Input price per 1M tokens	$0.95	$2.00
Output price per 1M tokens	$4.00	$10.00
Context window	262,144	No data
Image input	Yes	No data
Developer	Moonshot AI	OpenAI

Charts

Partial coverage. Terminus-2 – Kimi K2.6 (always on): 218 of 221 GS-QA2 tasks; Codex – GPT-6 Sol (high): 116 of 117 Earth-Bench V2 tasks; 138 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, Terminus-2 – Kimi K2.6 (always on) or Codex – GPT-6 Sol (high)?

Not comparable yet: Terminus-2 – Kimi K2.6 (always on) has partial coverage (218 of 221 GS-QA2 tasks) and Codex – GPT-6 Sol (high) has partial coverage (116 of 117 Earth-Bench V2 tasks; 138 of 150 GeoBenchX tasks), so it is not ranked. Terminus-2 – Kimi K2.6 (always on): 485 of 722 scheduled attempts. Codex – GPT-6 Sol (high): 460 of 722 scheduled attempts. Coverage is partial; scores can change as more attempts are decided.

Which is cheaper per task, Terminus-2 – Kimi K2.6 (always on) or Codex – GPT-6 Sol (high)?

Codex – GPT-6 Sol (high), at $0.080 against $0.089 for Terminus-2 – Kimi K2.6 (always on).

Which writes fewer output tokens per task, Terminus-2 – Kimi K2.6 (always on) or Codex – GPT-6 Sol (high)?

Codex – GPT-6 Sol (high), at 3k against 12k for Terminus-2 – Kimi K2.6 (always on).

Which has the larger context window, Terminus-2 – Kimi K2.6 (always on) or Codex – GPT-6 Sol (high)?

Not known yet: Codex – GPT-6 Sol (high) has no context window data.

More: Terminus-2 – Kimi K2.6 (always on) · Codex – GPT-6 Sol (high) · all comparisons