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

GLM-5.2 (max), Claude Code 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), Claude Code vs. Qwen 3.8 27B (xhigh), Qwen Code
Metric	GLM-5.2 (max), Claude Code	Qwen 3.8 27B (xhigh), Qwen Code
Geospatial Agent Index: Text only	73, not every task attempted yet	68
Geospatial Agent Index: Text + vision	No image input	Not run: Vision agents
Earth-Bench V2 score	71	68
GeoBenchX score	66	62
GS-QA2 score	83	75
Vision agents score	No image input	No data
GeoAgentBench score (Guided workflows; not in the Agent Index)	No data	94
GeoAnalystBench score (Guided workflows; not in the Agent Index)	No data	79
Raster and remote sensing score	80	76
Vector and overlay score	81	75
Climate and time series score	72	61
Spatial statistics and interpolation score	46	54
Mapping and cartography score	84	63
Recognising infeasible requests score	52	60
Cost per task	$0.34	$0.22
Time per task (at 100 output tokens/s)	6.3 min	4.7 min (estimated)
Tokens per attempt	617k	840k
Turns per attempt	15.4	17.6
Attempts scored	486	488
Input price per 1M tokens	$1.40	$0.45
Output price per 1M tokens	$4.40	$3.20
Context window	262,144	262,144
Image input	No	Yes
Developer	Z.ai	Alibaba (Qwen)

* Not every task attempted yet: the score may still change. What the asterisk means.

Charts

Not yet ranked. GLM-5.2 (max), Claude Code: 115 of 117 Earth-Bench V2 tasks attempted. Grey placeholders mark these entries in the index, cost and output token charts; their height is not a value. Bars appear once every task has an attempt. Results so far are in the side-by-side table.

Geospatial Agent Index

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

Data table: Geospatial Agent Index

Geospatial Agent Index
Model	Developer	Geospatial Agent Index	Status	Evaluations scored	Tasks attempted
Qwen 3.8 27B (xhigh), Qwen Code	Alibaba (Qwen)	68	Complete	3 of 3	488 of 488 tasks

Cost per task

Average cost per task (USD) · Lower is better

How to read this chart

Not shown: not yet ranked. 1 entries without an attempt on every index task. What is still to run is on the models page.

Data table: Cost per task

Cost per task
Model	Developer	Cost per task	Tasks attempted
Qwen 3.8 27B (xhigh), Qwen Code	Alibaba (Qwen)	$0.22	488 of 488 tasks

Output tokens per task

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

How to read this chart

Not shown: not yet ranked. 1 entries without an attempt on every index task. What is still to run is on the models page.

Data table: Output tokens per task

Output tokens per task
Model	Developer	Output tokens per task	Tasks attempted
Qwen 3.8 27B (xhigh), Qwen Code	Alibaba (Qwen)	28k	488 of 488 tasks

Frequently asked questions

Which scores higher on the Geospatial Agent Index, GLM-5.2 (max), Claude Code or Qwen 3.8 27B (xhigh), Qwen Code?

Not comparable yet: GLM-5.2 (max), Claude Code is not yet ranked (115 of 117 Earth-Bench V2 tasks attempted). GLM-5.2 (max), Claude Code: 486 of 488 tasks attempted. Qwen 3.8 27B (xhigh), Qwen Code: 488 of 488 tasks attempted. Scores can change as more tasks are attempted.

Which is cheaper per task, GLM-5.2 (max), Claude Code or Qwen 3.8 27B (xhigh), Qwen Code?

Qwen 3.8 27B (xhigh), Qwen Code, $0.22 to $0.34.

Which writes fewer output tokens per task, GLM-5.2 (max), Claude Code or Qwen 3.8 27B (xhigh), Qwen Code?

GLM-5.2 (max), Claude Code, 28k to 28k.

Which has the larger context window, GLM-5.2 (max), Claude Code or Qwen 3.8 27B (xhigh), Qwen Code?

They are level, at 262,144 tokens.

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