Claude Code – GLM-5.3 (max, default) vs. Codex – GPT-5.6 Terra (high): geospatial agent comparison

Comparison between Claude Code – GLM-5.3 (max, default) and Codex – GPT-5.6 Terra (high) 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 – GLM-5.3 (max, default) vs. Codex – GPT-5.6 Terra (high)
Metric	Claude Code – GLM-5.3 (max, default)	Codex – GPT-5.6 Terra (high)
Geospatial Agent Index	73*	81* (partial: attempt 1 on 131 of 488 tasks)
Earth-Bench V2 score	74	No data
GeoBenchX score	65	No data
GS-QA2 score	81	81
GeoAgentBench score (guided GIS workflow; outside the index)	98	No data
GeoAnalystBench score (guided GIS workflow; outside the index)	95	No data
Raster and remote sensing score	80	78
Vector and overlay score	81	83
Climate and time series score	76	No data
Spatial statistics and interpolation score	38	No data
Mapping and cartography score	75	No data
Recognising infeasible requests score	58	No data
Cost per task	$0.48	$0.053
Time per task	8.8 min	1.3 min
Tokens per attempt	489k	92k
Turns per attempt	13.4	7.4
Attempts decided	488	131
Input price per 1M tokens	$1.40	$2.00
Output price per 1M tokens	$4.40	$12.00
Context window	1,048,576	No data
Image input	No	No data
Developer	Z.ai	OpenAI

Charts

Partial coverage. Codex – GPT-5.6 Terra (high): Earth-Bench V2 incomplete; GeoBenchX incomplete; 131 of 221 GS-QA2 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

Attempts still to run

Data table: Geospatial Agent Index

Geospatial Agent Index
Model	Creator	Geospatial Agent Index	Range while attempts are pending	Benchmarks covered	Coverage (attempts)
Claude Code – GLM-5.3 (max, default)	Z.ai	73*	No pending judgements; coverage incomplete; 183 results awaiting review; 9 trace reviews declined by the judge	3 of 3	488 of 722 scheduled attempts (5 excluded)

Cost per task

Average cost per task (USD) · Lower is better

Attempts still to run

How to read this chart

Not shown: partial coverage. 1 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)
Claude Code – GLM-5.3 (max, default)	Z.ai	$0.48	488 of 722 scheduled attempts (5 excluded)

Output tokens per task

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

Attempts still to run

How to read this chart

Not shown: partial coverage. 1 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)
Claude Code – GLM-5.3 (max, default)	Z.ai	15k	488 of 722 scheduled attempts (5 excluded)

Frequently asked questions

Which scores higher on the Geospatial Agent Index, Claude Code – GLM-5.3 (max, default) or Codex – GPT-5.6 Terra (high)?

Not comparable yet: Codex – GPT-5.6 Terra (high) has partial coverage (Earth-Bench V2 incomplete; GeoBenchX incomplete; 131 of 221 GS-QA2 tasks), so it is not ranked. Claude Code – GLM-5.3 (max, default): 488 of 722 scheduled attempts (5 excluded). Codex – GPT-5.6 Terra (high): 131 of 572 scheduled attempts. Coverage is partial; scores can change as more attempts are decided.

Which is cheaper per task, Claude Code – GLM-5.3 (max, default) or Codex – GPT-5.6 Terra (high)?

Codex – GPT-5.6 Terra (high), at $0.053 against $0.48 for Claude Code – GLM-5.3 (max, default).

Which writes fewer output tokens per task, Claude Code – GLM-5.3 (max, default) or Codex – GPT-5.6 Terra (high)?

Codex – GPT-5.6 Terra (high), at 1k against 15k for Claude Code – GLM-5.3 (max, default).

Which has the larger context window, Claude Code – GLM-5.3 (max, default) or Codex – GPT-5.6 Terra (high)?

Not known yet: Codex – GPT-5.6 Terra (high) has no context window data.

More: Claude Code – GLM-5.3 (max, default) · Codex – GPT-5.6 Terra (high) · all comparisons