Axis Spatial

Search modelsMenu

Nemotron 3 Nano Omni 30B (Reasoning) vs Claude Opus 5 (high)

Results updated 15 November 2026 · In Terminus-2 and Claude Code, so part of any gap can come from the harness · How we benchmark

Verdict

Nemotron 3 Nano Omni 30B (Reasoning)21Geospatial Agent Index · rank 88 of 92
Claude Opus 5 (high)80Geospatial Agent Index · rank 10 of 92

Claude Opus 5 (high) clearly beats Nemotron 3 Nano Omni 30B (Reasoning) by 59 points, and is faster.

Cost per task
No data vs $0.32
Time per task
4.6 min vs 1.1 min
Tokens per attempt
564k vs 173k
Image input
Yes vs Yes

Claude Opus 5 (high), Claude Code scores higher than Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 on all 3 evaluations scored for both.

Where they differ most

By kind of work, share of tasks passed, 0 to 100 · Higher is better

Kind of workNemotron 3 Nano Omni 30B (Reasoning)Claude Opus 5 (high)
GIS analysis788
Remote sensing and raster analysis1785
Making maps1682
Climate and time series2880
World map of the places the tasks are setWhere the tasks are set: each kind of work links to its tasks and scores
Leaderboard

Charts for each: Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 · Claude Opus 5 (high), Claude Code.

Side by side

Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 vs. Claude Opus 5 (high), Claude Code
MetricNemotron 3 Nano Omni 30B (Reasoning), Terminus-2Claude Opus 5 (high), Claude Code
Geospatial Agent Index: Text only2180
Geospatial Agent Index: Text + visionNo data81
Earth-Bench V2 score3174
GeoBenchX score2375
GS-QA2 score991
Vision agents scoreNo data84
GeoAgentBench score (Guided workflows; not in the Agent Index)28100
GeoAnalystBench score (Guided workflows; not in the Agent Index)1689
OGD4All score (standalone; in no index)4795
GeoNatureAgent score (standalone; in no index)17100
MultiGlobeQA score (standalone; in no index)4879
GeoAgentEval score (standalone; in no index)4896
ImpactMesh score (standalone; in no index)No dataNo data
OpenEarthAgent Maps score (standalone; in no index)3399
OpenEarthAgent Sentinel-2 score (standalone; in no index)52100
GS-QA2 Full score (standalone; in no index)No data86
Place and coordinate tasks score (standalone; in no index)60No data
InSAR Regional Snow Mapping score (standalone; in no index)33100
DLPU phase unwrapping score (standalone; in no index)No dataNo data
Remote sensing and raster analysis score1785
Change detection score2868
GIS analysis score788
Climate and time series score2880
Making maps score1682
Cost per taskNo data$0.32
Time per task (at 100 output tokens/s)4.6 min1.1 min
Tokens per attempt564k173k
Turns per attempt30.27.2
Input price per 1M tokensNo data$5.00
Output price per 1M tokensNo data$25.00
Context window256,0001,000,000
Image inputYesYes
DeveloperNVIDIAAnthropic

Frequently asked questions

Which scores higher on the Geospatial Agent Index, Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 or Claude Opus 5 (high), Claude Code?
Claude Opus 5 (high), Claude Code, 80 to 21.
Which is cheaper per task, Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 or Claude Opus 5 (high), Claude Code?
Not known yet: Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 has no cost data.
Which writes fewer output tokens per task, Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 or Claude Opus 5 (high), Claude Code?
Claude Opus 5 (high), Claude Code, 4k to 21k.
Which has the larger context window, Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 or Claude Opus 5 (high), Claude Code?
Claude Opus 5 (high), Claude Code, 1,000,000 tokens to 256,000 tokens.