Axis Spatial

Claude Code – GLM-5.2 (max) vs. Codex – DeepSeek V4 Pro (high): geospatial agent comparison

Comparison between Claude Code – GLM-5.2 (max) and Codex – DeepSeek V4 Pro (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-5.2 (max) vs. Codex – DeepSeek V4 Pro (high)
MetricClaude Code – GLM-5.2 (max)Codex – DeepSeek V4 Pro (high)
Geospatial Agent Index53* (partial: attempt 1 on 200 of 565 tasks)40* (partial: attempt 1 on 48 of 565 tasks)
Earth-Bench V2 score6350
GeoBenchX score4229
GS-QA2 scoreNo dataNo data
GeoAgentBench score (guided GIS workflow; outside the index)No dataNo data
GeoAnalystBench score (guided GIS workflow; outside the index)No dataNo data
Raster and remote sensing score6255
Vector and overlay score40No data
Networks and routing scoreNo dataNo data
Climate and time series score670
Spatial statistics and interpolation scoreNo data0
Mapping and cartography scoreNo dataNo data
Recognising infeasible requests score4126
Cost per task$0.437$0.745
Time per task7.1 min12.8 min
Tokens per attempt827k852k
Turns per attempt18.724.6
Attempts decided12032
Input price per 1M tokens$1.40$1.32
Output price per 1M tokens$4.40$3.96
Context window262,1441,048,576
Image inputNoNo
DeveloperZ.aiDeepSeek

Charts

Partial coverage. Claude Code – GLM-5.2 (max): 35 of 147 Earth-Bench V2 tasks; 165 of 173 GeoBenchX tasks; GS-QA2 incomplete; Codex – DeepSeek V4 Pro (high): 8 of 147 Earth-Bench V2 tasks; 40 of 173 GeoBenchX tasks; GS-QA2 incomplete. 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

Geospatial Agent IndexClaude Code – GLM-5.2 (max): Partial; Codex – DeepSeek V4 Pro (high): Partial0255075100PartialGLM-5.2 (max)PartialDeepSeek V4 Pro (high)
Data table: Geospatial Agent Index
Geospatial Agent Index
ModelCreatorGeospatial Agent IndexRange while attempts are pendingBenchmarks coveredCoverage (attempts)

Cost per task

Average cost per task (USD) · Lower is better

Cost per taskClaude Code – GLM-5.2 (max): Partial; Codex – DeepSeek V4 Pro (high): Partial$0$0.250$0.500$0.750$1.00PartialGLM-5.2 (max)PartialDeepSeek V4 Pro (high)
How to read this chart

Not shown: partial coverage. Claude Code – GLM-5.2 (max), Codex – DeepSeek V4 Pro (high).

Data table: Cost per task
Cost per task
ModelCreatorCost per taskCoverage (attempts)

Output tokens per task

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

Output tokens per taskClaude Code – GLM-5.2 (max): Partial; Codex – DeepSeek V4 Pro (high): Partial00011PartialGLM-5.2 (max)PartialDeepSeek V4 Pro (high)
How to read this chart

Not shown: partial coverage. Claude Code – GLM-5.2 (max), Codex – DeepSeek V4 Pro (high).

Data table: Output tokens per task
Output tokens per task
ModelCreatorOutput tokens per taskCoverage (attempts)

Frequently asked questions

Which scores higher on the Geospatial Agent Index, Claude Code – GLM-5.2 (max) or Codex – DeepSeek V4 Pro (high)?
Not comparable yet: Claude Code – GLM-5.2 (max) has partial coverage (35 of 147 Earth-Bench V2 tasks; 165 of 173 GeoBenchX tasks; GS-QA2 incomplete) and Codex – DeepSeek V4 Pro (high) has partial coverage (8 of 147 Earth-Bench V2 tasks; 40 of 173 GeoBenchX tasks; GS-QA2 incomplete), so it is not ranked. Claude Code – GLM-5.2 (max): 120 of 614 planned attempts (9 excluded). Codex – DeepSeek V4 Pro (high): 32 of 614 planned attempts. Coverage is partial; scores can change as more attempts are decided.
Which is cheaper per task, Claude Code – GLM-5.2 (max) or Codex – DeepSeek V4 Pro (high)?
Claude Code – GLM-5.2 (max), at $0.437 against $0.745 for Codex – DeepSeek V4 Pro (high).
Which writes fewer output tokens per task, Claude Code – GLM-5.2 (max) or Codex – DeepSeek V4 Pro (high)?
Codex – DeepSeek V4 Pro (high), at 19k against 34k for Claude Code – GLM-5.2 (max).
Which has the larger context window, Claude Code – GLM-5.2 (max) or Codex – DeepSeek V4 Pro (high)?
Codex – DeepSeek V4 Pro (high), at 1,048,576 tokens against 262,144 tokens for Claude Code – GLM-5.2 (max).

More: Claude Code – GLM-5.2 (max) · Codex – DeepSeek V4 Pro (high) · all comparisons