Axis Spatial

Terminus-2 – GLM-5.2 (none) vs. Terminus-2 – GLM-5.3 (low): geospatial agent comparison

Comparison between Terminus-2 – GLM-5.2 (none) and Terminus-2 – GLM-5.3 (low) 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 – GLM-5.2 (none) vs. Terminus-2 – GLM-5.3 (low)
MetricTerminus-2 – GLM-5.2 (none)Terminus-2 – GLM-5.3 (low)
Geospatial Agent Index69* (partial: 1.4 of 3 attempts)71* (partial: 0.9 of 3 attempts)
GeoAgentBench score9495
GeoBenchX score5151
Earth-Bench score5138
GeoAnalystBench score79100
Raster and remote sensing score7365
Vector and overlay score6979
Networks and routing score100100
Climate and time series score5860
Spatial statistics and interpolation score5850
Mapping and cartography score7572
Recognising infeasible requests score3338
Cost per task$0.103$0.250
Time per task2.9 min6.7 min
Tokens per attempt169k205k
Turns per attempt11.310.7
Attempts decided356238
Input price per 1M tokens$1.40$1.40
Output price per 1M tokens$4.40$4.40
Context window262,1441,048,576
Image inputNoNo
DeveloperZ.aiZ.ai

Charts

Geospatial Agent Index

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

  • Partial coverage
Geospatial Agent IndexTerminus-2 – GLM-5.3 (low): 71* (partial coverage); Terminus-2 – GLM-5.2 (none): 69* (partial coverage)025507510071*GLM-5.3 (low)69*GLM-5.2 (none)
Data table: Geospatial Agent Index
Geospatial Agent Index
ModelCreatorGeospatial Agent IndexRange while attempts are pendingBenchmarks coveredCoverage (attempts)
Terminus-2 – GLM-5.3 (low)Z.ai71* (partial: 0.9 of 3 attempts)61 to 724 of 4238 of 290 planned attempts
Terminus-2 – GLM-5.2 (none)Z.ai69* (partial: 1.4 of 3 attempts)62 to 724 of 4356 of 870 planned attempts

Cost per task

Average cost per task (USD) · Lower is better

  • Partial coverage
Cost per taskTerminus-2 – GLM-5.2 (none): $0.103 (partial coverage); Terminus-2 – GLM-5.3 (low): $0.250 (partial coverage)$0$0.062$0.125$0.188$0.250$0.103GLM-5.2 (none)$0.250GLM-5.3 (low)
Data table: Cost per task
Cost per task
ModelCreatorCost per taskCoverage (attempts)
Terminus-2 – GLM-5.2 (none)Z.ai$0.103408 of 870 planned attempts
Terminus-2 – GLM-5.3 (low)Z.ai$0.250258 of 290 planned attempts

Time per task

Average agent wall time per task · Lower is better

  • Partial coverage
Time per taskTerminus-2 – GLM-5.2 (none): 2.9 min (partial coverage); Terminus-2 – GLM-5.3 (low): 6.7 min (partial coverage)0m2m3m5m7m2.9 minGLM-5.2 (none)6.7 minGLM-5.3 (low)
Data table: Time per task
Time per task
ModelCreatorTime per taskCoverage (attempts)
Terminus-2 – GLM-5.2 (none)Z.ai2.9 min408 of 870 planned attempts
Terminus-2 – GLM-5.3 (low)Z.ai6.7 min258 of 290 planned attempts

Frequently asked questions

Which scores higher on the Geospatial Agent Index, Terminus-2 – GLM-5.2 (none) or Terminus-2 – GLM-5.3 (low)?
Terminus-2 – GLM-5.3 (low), at 71 against 69 for Terminus-2 – GLM-5.2 (none). Terminus-2 – GLM-5.2 (none): 356 of 870 planned attempts. Terminus-2 – GLM-5.3 (low): 238 of 290 planned attempts. Coverage is partial; scores can change as more attempts are decided.
Which is cheaper per task, Terminus-2 – GLM-5.2 (none) or Terminus-2 – GLM-5.3 (low)?
Terminus-2 – GLM-5.2 (none), at $0.103 against $0.250 for Terminus-2 – GLM-5.3 (low).
Which is faster per task, Terminus-2 – GLM-5.2 (none) or Terminus-2 – GLM-5.3 (low)?
Terminus-2 – GLM-5.2 (none), at 2.9 min against 6.7 min for Terminus-2 – GLM-5.3 (low).
Which has the larger context window, Terminus-2 – GLM-5.2 (none) or Terminus-2 – GLM-5.3 (low)?
Terminus-2 – GLM-5.3 (low), at 1,048,576 tokens against 262,144 tokens for Terminus-2 – GLM-5.2 (none).

More: Terminus-2 – GLM-5.2 (none) · Terminus-2 – GLM-5.3 (low) · all comparisons