Axis Spatial

Nemotron 3 Ultra (high), Terminus-2 vs. Qwen 3.8 27B (xhigh), Qwen Code

Nemotron 3 Ultra (high), 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

Nemotron 3 Ultra (high), Terminus-2 vs. Qwen 3.8 27B (xhigh), Qwen Code
MetricNemotron 3 Ultra (high), Terminus-2Qwen 3.8 27B (xhigh), Qwen Code
Geospatial Agent Index: Text only6868
Geospatial Agent Index: Text + visionNo dataNo data
Earth-Bench V2 score6868
GeoBenchX score6562
GS-QA2 score7075
Vision agents scoreNo dataNo data
GeoAgentBench score (Guided workflows; not in the Agent Index)9694
GeoAnalystBench score (Guided workflows; not in the Agent Index)6379
OGD4All score (standalone; in no index)No dataNo data
GeoNatureAgent score (standalone; in no index)No dataNo data
MultiGlobeQA score (standalone; in no index)No dataNo data
GeoAgentEval score (standalone; in no index)No dataNo data
Remote sensing and raster analysis score7076
GIS analysis score7075
Climate and time series score6861
Spatial statistics score3154
Making maps score7063
Recognising infeasible requests score6860
Cost per task$0.069$0.22
Time per task (at 100 output tokens/s)2.4 min4.7 min (estimated)
Tokens per attempt148k840k
Turns per attempt12.117.6
Input price per 1M tokens$0.50$0.45
Output price per 1M tokens$2.20$3.20
Context window262,144262,144
Image inputNoYes
DeveloperNVIDIAAlibaba (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: Qwen 3.8 27B (xhigh), Qwen Code: 68; Nemotron 3 Ultra (high), Terminus-2: 68025507510068Qwen 3.8 27B (xhigh)68Nemotron 3 Ultra (high)
Data table: Geospatial Agent Index
Geospatial Agent Index
ModelDeveloperGeospatial Agent Index
Qwen 3.8 27B (xhigh), Qwen CodeAlibaba (Qwen)68
Nemotron 3 Ultra (high), Terminus-2NVIDIA68

Cost per task

Average cost per task (USD) · Lower is better

Cost per taskCost per task in US dollars: Nemotron 3 Ultra (high), Terminus-2: $0.069; Qwen 3.8 27B (xhigh), Qwen Code: $0.22$0$0.10$0.20$0.30$0.40$0.069Nemotron 3 Ultra (high)$0.22Qwen 3.8 27B (xhigh)
Data table: Cost per task
Cost per task
ModelDeveloperCost per task
Nemotron 3 Ultra (high), Terminus-2NVIDIA$0.069
Qwen 3.8 27B (xhigh), Qwen CodeAlibaba (Qwen)$0.22

Output tokens per task

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

Output tokens per taskOutput tokens per task: Nemotron 3 Ultra (high), Terminus-2: 10k; Qwen 3.8 27B (xhigh), Qwen Code: 28k010k20k30k40k10kNemotron 3 Ultra (high)28kQwen 3.8 27B (xhigh)
Data table: Output tokens per task
Output tokens per task
ModelDeveloperOutput tokens per task
Nemotron 3 Ultra (high), Terminus-2NVIDIA10k
Qwen 3.8 27B (xhigh), Qwen CodeAlibaba (Qwen)28k

Frequently asked questions

Which scores higher on the Geospatial Agent Index, Nemotron 3 Ultra (high), Terminus-2 or Qwen 3.8 27B (xhigh), Qwen Code?
Qwen 3.8 27B (xhigh), Qwen Code, 68 to 68.
Which is cheaper per task, Nemotron 3 Ultra (high), Terminus-2 or Qwen 3.8 27B (xhigh), Qwen Code?
Nemotron 3 Ultra (high), Terminus-2, $0.069 to $0.22.
Which writes fewer output tokens per task, Nemotron 3 Ultra (high), Terminus-2 or Qwen 3.8 27B (xhigh), Qwen Code?
Nemotron 3 Ultra (high), Terminus-2, 10k to 28k.
Which has the larger context window, Nemotron 3 Ultra (high), Terminus-2 or Qwen 3.8 27B (xhigh), Qwen Code?
They are level, at 262,144 tokens.

More: Nemotron 3 Ultra (high), Terminus-2 · Qwen 3.8 27B (xhigh), Qwen Code · All comparisons