DeepSeek V4 Pro (high), Terminus-2 vs. Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2
DeepSeek V4 Pro (high), Terminus-2 against Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2: Geospatial Agent Index, each evaluation, cost, time and tokens, on the same tasks and limits. Methodology.
Side by side
| Metric | DeepSeek V4 Pro (high), Terminus-2 | Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 |
|---|---|---|
| Geospatial Agent Index: Text only | 70 | 20 |
| Geospatial Agent Index: Text + vision | No image input | No data |
| Earth-Bench V2 score | 68 | 29 |
| GeoBenchX score | 65 | 21 |
| GS-QA2 score | 77 | 9 |
| Vision agents score | No image input | No data |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 96 | 26 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 100 | 16 |
| OGD4All score (standalone; in no index) | No data | No data |
| GeoNatureAgent score (standalone; in no index) | No data | No data |
| MultiGlobeQA score (standalone; in no index) | No data | No data |
| GeoAgentEval score (standalone; in no index) | No data | No data |
| Remote sensing and raster analysis score | 77 | 17 |
| GIS analysis score | 73 | 6 |
| Climate and time series score | 66 | 27 |
| Spatial statistics score | 23 | 0 |
| Making maps score | 81 | 16 |
| Recognising infeasible requests score | 62 | 33 |
| Cost per task | $0.30 | $0.13 |
| Time per task (at 100 output tokens/s) | 4.0 min | 4.9 min |
| Tokens per attempt | 242k | 589k |
| Turns per attempt | 13.9 | 30.7 |
| Input price per 1M tokens | $1.32 | $0.20 |
| Output price per 1M tokens | $3.96 | $0.80 |
| Context window | 1,048,576 | 256,000 |
| Image input | No | Yes |
| Developer | DeepSeek | NVIDIA |
Charts
Geospatial Agent Index
Equal average of Earth-Bench V2, GeoBenchX and GS-QA2, 0 to 100 · Higher is better
Data table: Geospatial Agent Index
| Model | Developer | Geospatial Agent Index |
|---|---|---|
| DeepSeek V4 Pro (high), Terminus-2 | DeepSeek | 70 |
| Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 | NVIDIA | 20 |
Cost per task
Average cost per task (USD) · Lower is better
Data table: Cost per task
| Model | Developer | Cost per task |
|---|---|---|
| Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 | NVIDIA | $0.13 |
| DeepSeek V4 Pro (high), Terminus-2 | DeepSeek | $0.30 |
Output tokens per task
Average output tokens per attempt, reasoning included · Lower is better
Data table: Output tokens per task
| Model | Developer | Output tokens per task |
|---|---|---|
| DeepSeek V4 Pro (high), Terminus-2 | DeepSeek | 20k |
| Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 | NVIDIA | 22k |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, DeepSeek V4 Pro (high), Terminus-2 or Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2?
DeepSeek V4 Pro (high), Terminus-2, 70 to 20.
Which is cheaper per task, DeepSeek V4 Pro (high), Terminus-2 or Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2?
Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2, $0.13 to $0.30.
Which writes fewer output tokens per task, DeepSeek V4 Pro (high), Terminus-2 or Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2?
DeepSeek V4 Pro (high), Terminus-2, 20k to 22k.
Which has the larger context window, DeepSeek V4 Pro (high), Terminus-2 or Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2?
DeepSeek V4 Pro (high), Terminus-2, 1,048,576 tokens to 256,000 tokens.
More: DeepSeek V4 Pro (high), Terminus-2 · Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 · All comparisons