Axis Spatial

Nemotron 3 Ultra (high), Terminus-2 vs. Claude Opus 4.7 (high), Claude Code

Nemotron 3 Ultra (high), Terminus-2 against Claude Opus 4.7 (high), Claude 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. Claude Opus 4.7 (high), Claude Code
MetricNemotron 3 Ultra (high), Terminus-2Claude Opus 4.7 (high), Claude Code
Geospatial Agent Index: Text only6874
Geospatial Agent Index: Text + visionNo data74
Earth-Bench V2 score6869
GeoBenchX score6573
GS-QA2 score7079
Vision agents scoreNo data77
GeoAgentBench score (Guided workflows; not in the Agent Index)9696
GeoAnalystBench score (Guided workflows; not in the Agent Index)63100
OGD4All score (standalone; in no index)No data95
GeoNatureAgent score (standalone; in no index)No data100
MultiGlobeQA score (standalone; in no index)No data77
GeoAgentEval score (standalone; in no index)No data92
Remote sensing and raster analysis score7079
GIS analysis score7075
Climate and time series score6871
Spatial statistics score3138
Making maps score7082
Recognising infeasible requests score6870
Cost per task$0.069$0.34
Time per task (at 100 output tokens/s)2.4 min0.9 min
Tokens per attempt148k245k
Turns per attempt12.17.0
Input price per 1M tokens$0.50$5.00
Output price per 1M tokens$2.20$25.00
Context window262,1441,000,000
Image inputNoYes
DeveloperNVIDIAAnthropic

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: Claude Opus 4.7 (high), Claude Code: 74; Nemotron 3 Ultra (high), Terminus-2: 68025507510074Claude Opus 4.7 (high)68Nemotron 3 Ultra (high)
Data table: Geospatial Agent Index
Geospatial Agent Index
ModelDeveloperGeospatial Agent Index
Claude Opus 4.7 (high), Claude CodeAnthropic74
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; Claude Opus 4.7 (high), Claude Code: $0.34$0$0.10$0.20$0.30$0.40$0.069Nemotron 3 Ultra (high)$0.34Claude Opus 4.7 (high)
Data table: Cost per task
Cost per task
ModelDeveloperCost per task
Nemotron 3 Ultra (high), Terminus-2NVIDIA$0.069
Claude Opus 4.7 (high), Claude CodeAnthropic$0.34

Output tokens per task

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

Output tokens per taskOutput tokens per task: Claude Opus 4.7 (high), Claude Code: 4k; Nemotron 3 Ultra (high), Terminus-2: 10k02k5k8k10k4kClaude Opus 4.7 (high)10kNemotron 3 Ultra (high)
Data table: Output tokens per task
Output tokens per task
ModelDeveloperOutput tokens per task
Claude Opus 4.7 (high), Claude CodeAnthropic4k
Nemotron 3 Ultra (high), Terminus-2NVIDIA10k

Frequently asked questions

Which scores higher on the Geospatial Agent Index, Nemotron 3 Ultra (high), Terminus-2 or Claude Opus 4.7 (high), Claude Code?
Claude Opus 4.7 (high), Claude Code, 74 to 68.
Which is cheaper per task, Nemotron 3 Ultra (high), Terminus-2 or Claude Opus 4.7 (high), Claude Code?
Nemotron 3 Ultra (high), Terminus-2, $0.069 to $0.34.
Which writes fewer output tokens per task, Nemotron 3 Ultra (high), Terminus-2 or Claude Opus 4.7 (high), Claude Code?
Claude Opus 4.7 (high), Claude Code, 4k to 10k.
Which has the larger context window, Nemotron 3 Ultra (high), Terminus-2 or Claude Opus 4.7 (high), Claude Code?
Claude Opus 4.7 (high), Claude Code, 1,000,000 tokens to 262,144 tokens.

More: Nemotron 3 Ultra (high), Terminus-2 · Claude Opus 4.7 (high), Claude Code · All comparisons