Axis Spatial

GLM-5.2 (max), Terminus-2 vs. Qwen 3.8 27B (xhigh), Qwen Code

GLM-5.2 (max), Terminus-2 against Qwen 3.8 27B (xhigh), Qwen Code: Geospatial Agent Index, each evaluation, cost, time and tokens, on the same tasks and limits. The harnesses differ, so part of any gap can come from the harness. Methodology.

Side by side

GLM-5.2 (max), Terminus-2 vs. Qwen 3.8 27B (xhigh), Qwen Code
MetricGLM-5.2 (max), Terminus-2Qwen 3.8 27B (xhigh), Qwen Code
Geospatial Agent Index: Text only7268
Geospatial Agent Index: Text + visionNo image inputNot run: Vision agents
Earth-Bench V2 score6868
GeoBenchX score6762
GS-QA2 score8075
Vision agents scoreNo image inputNo data
GeoAgentBench score (Guided workflows; not in the Agent Index)9494
GeoAnalystBench score (Guided workflows; not in the Agent Index)9579
Raster and remote sensing score7676
Vector and overlay score8275
Climate and time series score7161
Spatial statistics and interpolation score6254
Mapping and cartography score8263
Recognising infeasible requests score4860
Cost per task$0.32$0.22
Time per task (at 100 output tokens/s)6.1 min4.7 min (estimated)
Tokens per attempt416k840k
Turns per attempt13.617.6
Attempts scored488488
Input price per 1M tokens$1.40$0.45
Output price per 1M tokens$4.40$3.20
Context window262,144262,144
Image inputNoYes
DeveloperZ.aiAlibaba (Qwen)

Charts

Geospatial Agent Index

Equal average of Earth-Bench V2, GeoBenchX and GS-QA2, 0 to 100 · Higher is better

Geospatial Agent IndexScores 0 to 100: GLM-5.2 (max), Terminus-2: 72; Qwen 3.8 27B (xhigh), Qwen Code: 68025507510072GLM-5.2 (max)68Qwen 3.8 27B (xhigh)
Data table: Geospatial Agent Index
Geospatial Agent Index
ModelDeveloperGeospatial Agent IndexStatusEvaluations scoredTasks attempted
GLM-5.2 (max), Terminus-2Z.ai72Complete3 of 3488 of 488 tasks
Qwen 3.8 27B (xhigh), Qwen CodeAlibaba (Qwen)68Complete3 of 3488 of 488 tasks

Cost per task

Average cost per task (USD) · Lower is better

Cost per taskCost per task in US dollars: Qwen 3.8 27B (xhigh), Qwen Code: $0.22; GLM-5.2 (max), Terminus-2: $0.32$0$0.10$0.20$0.30$0.40$0.22Qwen 3.8 27B (xhigh)$0.32GLM-5.2 (max)
Data table: Cost per task
Cost per task
ModelDeveloperCost per taskTasks attempted
Qwen 3.8 27B (xhigh), Qwen CodeAlibaba (Qwen)$0.22488 of 488 tasks
GLM-5.2 (max), Terminus-2Z.ai$0.32488 of 488 tasks

Output tokens per task

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

Output tokens per taskOutput tokens per task: Qwen 3.8 27B (xhigh), Qwen Code: 28k; GLM-5.2 (max), Terminus-2: 30k010k20k30k40k28kQwen 3.8 27B (xhigh)30kGLM-5.2 (max)
Data table: Output tokens per task
Output tokens per task
ModelDeveloperOutput tokens per taskTasks attempted
Qwen 3.8 27B (xhigh), Qwen CodeAlibaba (Qwen)28k488 of 488 tasks
GLM-5.2 (max), Terminus-2Z.ai30k488 of 488 tasks

Frequently asked questions

Which scores higher on the Geospatial Agent Index, GLM-5.2 (max), Terminus-2 or Qwen 3.8 27B (xhigh), Qwen Code?
GLM-5.2 (max), Terminus-2, 72 to 68. Both have attempted all 488 tasks.
Which is cheaper per task, GLM-5.2 (max), Terminus-2 or Qwen 3.8 27B (xhigh), Qwen Code?
Qwen 3.8 27B (xhigh), Qwen Code, $0.22 to $0.32.
Which writes fewer output tokens per task, GLM-5.2 (max), Terminus-2 or Qwen 3.8 27B (xhigh), Qwen Code?
Qwen 3.8 27B (xhigh), Qwen Code, 28k to 30k.
Which has the larger context window, GLM-5.2 (max), Terminus-2 or Qwen 3.8 27B (xhigh), Qwen Code?
They are level, at 262,144 tokens.

More: GLM-5.2 (max), Terminus-2 · Qwen 3.8 27B (xhigh), Qwen Code · All comparisons