gpt-oss-20b (high), Terminus-2 vs. Mistral Large 3 (Non-reasoning), Terminus-2
gpt-oss-20b (high), Terminus-2 against Mistral Large 3 (Non-reasoning), Terminus-2: Geospatial Agent Index, each evaluation, cost, time and tokens, on the same tasks and limits. Methodology.
Side by side
| Metric | gpt-oss-20b (high), Terminus-2 | Mistral Large 3 (Non-reasoning), Terminus-2 |
|---|---|---|
| Geospatial Agent Index: Text only | 33 | 33, not every task attempted yet |
| Geospatial Agent Index: Text + vision | No image input | Still to run: GS-QA2, Vision agents |
| Earth-Bench V2 score | 32 | 31 |
| GeoBenchX score | 38 | 29 |
| GS-QA2 score | 30 | 39 |
| Vision agents score | No image input | No data |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 53 | 46 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 37 | 47 |
| Raster and remote sensing score | 33 | 29 |
| Vector and overlay score | 26 | 41 |
| Climate and time series score | 31 | 34 |
| Spatial statistics and interpolation score | 0 | 8 |
| Mapping and cartography score | 21 | 33 |
| Recognising infeasible requests score | 67 | 33 |
| Cost per task | $0.017 | $0.026 |
| Time per task (at 100 output tokens/s) | 7.2 min | 1.1 min |
| Tokens per attempt | 67k | 43k |
| Turns per attempt | 5.2 | 5.2 |
| Attempts scored | 488 | 487 |
| Input price per 1M tokens | $0.20 | $0.50 |
| Output price per 1M tokens | $0.30 | $1.50 |
| Context window | 128,000 | 256,000 |
| Image input | No | Yes |
| Developer | OpenAI | Mistral AI |
* Not every task attempted yet: the score may still change. What the asterisk means.
Charts
Not yet ranked. Mistral Large 3 (Non-reasoning), Terminus-2: 220 of 221 GS-QA2 tasks attempted. Grey placeholders mark these entries in the index, cost and output token charts; their height is not a value. Bars appear once every task has an attempt. Results so far are in the side-by-side table.
Geospatial Agent Index
Equal average of Earth-Bench V2, GeoBenchX and GS-QA2, 0 to 100 · Higher is better
Top eight. Every model is in the table.
Data table: Geospatial Agent Index
| Model | Developer | Geospatial Agent Index | Status | Evaluations scored | Tasks attempted |
|---|---|---|---|---|---|
| gpt-oss-20b (high), Terminus-2 | OpenAI | 33 | Complete | 3 of 3 | 488 of 488 tasks (6 attempts not counted: provider or infrastructure error) |
Cost per task
Average cost per task (USD) · Lower is better
Top eight. Every model is in the table.
How to read this chart
Not shown: not yet ranked. 1 entries without an attempt on every index task. What is still to run is on the models page.
Data table: Cost per task
| Model | Developer | Cost per task | Tasks attempted |
|---|---|---|---|
| gpt-oss-20b (high), Terminus-2 | OpenAI | $0.017 | 488 of 488 tasks (6 attempts not counted: provider or infrastructure error) |
Output tokens per task
Average output tokens per attempt, reasoning included · Lower is better
Top eight. Every model is in the table.
How to read this chart
Not shown: not yet ranked. 1 entries without an attempt on every index task. What is still to run is on the models page.
Data table: Output tokens per task
| Model | Developer | Output tokens per task | Tasks attempted |
|---|---|---|---|
| gpt-oss-20b (high), Terminus-2 | OpenAI | 40k | 488 of 488 tasks (6 attempts not counted: provider or infrastructure error) |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, gpt-oss-20b (high), Terminus-2 or Mistral Large 3 (Non-reasoning), Terminus-2?
Which is cheaper per task, gpt-oss-20b (high), Terminus-2 or Mistral Large 3 (Non-reasoning), Terminus-2?
Which writes fewer output tokens per task, gpt-oss-20b (high), Terminus-2 or Mistral Large 3 (Non-reasoning), Terminus-2?
Which has the larger context window, gpt-oss-20b (high), Terminus-2 or Mistral Large 3 (Non-reasoning), Terminus-2?
More: gpt-oss-20b (high), Terminus-2 · Mistral Large 3 (Non-reasoning), Terminus-2 · All comparisons