Terminus-2 – Gemma 4 26B A4B (Reasoning) vs. Terminus-2 – gpt-oss-120b (high): geospatial agent comparison
Comparison between Terminus-2 – Gemma 4 26B A4B (Reasoning) and Terminus-2 – gpt-oss-120b (high) on the Geospatial Agent Index, each benchmark, cost, time and token use, from the same tasks, agent and limits. Methodology.
Side by side
| Metric | Terminus-2 – Gemma 4 26B A4B (Reasoning) | Terminus-2 – gpt-oss-120b (high) |
|---|---|---|
| Geospatial Agent Index | 60* | 53* |
| GeoAgentBench score | 86 | 70 |
| GeoBenchX score | 55 | 42 |
| Earth-Bench score | 30 | 46 |
| GeoAnalystBench score | 68 | 53 |
| Raster and remote sensing score | 51 | 52 |
| Vector and overlay score | 79 | 56 |
| Networks and routing score | 100 | 100 |
| Climate and time series score | 48 | 62 |
| Spatial statistics and interpolation score | 75 | 30 |
| Mapping and cartography score | 83 | 43 |
| Recognising infeasible requests score | 44 | 47 |
| Cost per task | $0.012 | $0.026 |
| Time per task | 8.6 min | 4.6 min |
| Tokens per attempt | 89k | 59k |
| Turns per attempt | 8.7 | 5.3 |
| Attempts decided | 145 | 249 |
| Input price per 1M tokens | $0.10 | $0.35 |
| Output price per 1M tokens | $0.30 | $0.75 |
| Context window | 256,000 | 128,000 |
| Image input | Yes | No |
| Developer | OpenAI |
Charts
Axis Spatial Geospatial Agent Index
Equal-weight average of the evaluation scores, 0 to 100 · Higher is better
Data table: Axis Spatial Geospatial Agent Index
| Model | Creator | Geospatial Agent Index | Range while attempts are pending | Benchmarks covered | Coverage (attempts) |
|---|---|---|---|---|---|
| Terminus-2 – Gemma 4 26B A4B (Reasoning) | 60* | 60 to 60 | 4 of 4 | 145 of 870 planned attempts | |
| Terminus-2 – gpt-oss-120b (high) | OpenAI | 53* | No pending judgements; coverage incomplete | 4 of 4 | 249 of 870 planned attempts |
Cost per task
Average cost per task (USD) · Lower is better
Data table: Cost per task
| Model | Creator | Cost per task | Coverage (attempts) |
|---|---|---|---|
| Terminus-2 – Gemma 4 26B A4B (Reasoning) | $0.012 | 146 of 870 planned attempts | |
| Terminus-2 – gpt-oss-120b (high) | OpenAI | $0.026 | 249 of 870 planned attempts |
Time per task
Average agent wall time per task · Lower is better
Data table: Time per task
| Model | Creator | Time per task | Coverage (attempts) |
|---|---|---|---|
| Terminus-2 – Gemma 4 26B A4B (Reasoning) | 8.6 min | 146 of 870 planned attempts | |
| Terminus-2 – gpt-oss-120b (high) | OpenAI | 4.6 min | 249 of 870 planned attempts |
Frequently asked questions
- Which scores higher on the Geospatial Agent Index, Terminus-2 – Gemma 4 26B A4B (Reasoning) or Terminus-2 – gpt-oss-120b (high)?
- Terminus-2 – Gemma 4 26B A4B (Reasoning), at 60 against 53 for Terminus-2 – gpt-oss-120b (high). Terminus-2 – Gemma 4 26B A4B (Reasoning): 145 of 870 planned attempts. Terminus-2 – gpt-oss-120b (high): 249 of 870 planned attempts. Coverage is partial; scores can change as more attempts are decided.
- Which is cheaper per task, Terminus-2 – Gemma 4 26B A4B (Reasoning) or Terminus-2 – gpt-oss-120b (high)?
- Terminus-2 – Gemma 4 26B A4B (Reasoning), at $0.012 against $0.026 for Terminus-2 – gpt-oss-120b (high).
- Which is faster per task, Terminus-2 – Gemma 4 26B A4B (Reasoning) or Terminus-2 – gpt-oss-120b (high)?
- Terminus-2 – gpt-oss-120b (high), at 4.6 min against 8.6 min for Terminus-2 – Gemma 4 26B A4B (Reasoning).
- Which has the larger context window, Terminus-2 – Gemma 4 26B A4B (Reasoning) or Terminus-2 – gpt-oss-120b (high)?
- Terminus-2 – Gemma 4 26B A4B (Reasoning), at 256,000 tokens against 128,000 tokens for Terminus-2 – gpt-oss-120b (high).
More: Terminus-2 – Gemma 4 26B A4B (Reasoning) · Terminus-2 – gpt-oss-120b (high) · all comparisons