Terminus-2 – Llama 3.3 70B (Non-reasoning) vs. Codex – gpt-oss-20b (high): geospatial agent comparison
Comparison between Terminus-2 – Llama 3.3 70B (Non-reasoning) and Codex – gpt-oss-20b (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 – Llama 3.3 70B (Non-reasoning) | Codex – gpt-oss-20b (high) |
|---|---|---|
| Geospatial Agent Index | 8* | 4* (partial: attempt 1 on 173 of 565 tasks) |
| Earth-Bench V2 score | 7 | 0 |
| GeoBenchX score | 6 | 8 |
| GS-QA2 score | 10 | No data |
| GeoAgentBench score (guided GIS workflow; outside the index) | 8 | No data |
| GeoAnalystBench score (guided GIS workflow; outside the index) | 0 | No data |
| Raster and remote sensing score | 6 | 0 |
| Vector and overlay score | 11 | 25 |
| Networks and routing score | No data | No data |
| Climate and time series score | 10 | 0 |
| Spatial statistics and interpolation score | 0 | 0 |
| Mapping and cartography score | 2 | 4 |
| Recognising infeasible requests score | 13 | 12 |
| Cost per task | $0.022 | $0.140 |
| Time per task | 3.0 min | 8.0 min |
| Tokens per attempt | 59k | 683k |
| Turns per attempt | 14.0 | 29.5 |
| Attempts decided | 565 | 170 |
| Input price per 1M tokens | $0.29 | $0.20 |
| Output price per 1M tokens | $2.25 | $0.30 |
| Context window | 24,000 | 128,000 |
| Image input | No | No |
| Developer | Meta | OpenAI |
Charts
Partial coverage. Codex – gpt-oss-20b (high): 32 of 147 Earth-Bench V2 tasks; 141 of 173 GeoBenchX tasks; GS-QA2 incomplete. Grey placeholders identify these entries in the index, cost and output token charts; their height does not represent a value. Comparable bars appear when first-attempt coverage is complete. Their available results remain in the side-by-side table.
Geospatial Agent Index
Equal-weight average of the evaluation scores, 0 to 100 · Higher is better
First eight selected entries in metric order. Full results in table view.
Data table: Geospatial Agent Index
| Model | Creator | Geospatial Agent Index | Range while attempts are pending | Benchmarks covered | Coverage (attempts) |
|---|---|---|---|---|---|
| Terminus-2 – Llama 3.3 70B (Non-reasoning) | Meta | 8* | No pending judgements; coverage incomplete; 24 results awaiting review | 3 of 3 | 565 of 859 planned attempts |
Cost per task
Average cost per task (USD) · Lower is better
First eight selected entries in metric order. Full results in table view.
How to read this chart
Not shown: partial coverage. Codex – gpt-oss-20b (high).
Data table: Cost per task
| Model | Creator | Cost per task | Coverage (attempts) |
|---|---|---|---|
| Terminus-2 – Llama 3.3 70B (Non-reasoning) | Meta | $0.022 | 565 of 859 planned attempts |
Output tokens per task
Average output tokens per attempt, reasoning included · Lower is better
First eight selected entries in metric order. Full results in table view.
How to read this chart
Not shown: partial coverage. Codex – gpt-oss-20b (high).
Data table: Output tokens per task
| Model | Creator | Output tokens per task | Coverage (attempts) |
|---|---|---|---|
| Terminus-2 – Llama 3.3 70B (Non-reasoning) | Meta | 2k | 565 of 859 planned attempts |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, Terminus-2 – Llama 3.3 70B (Non-reasoning) or Codex – gpt-oss-20b (high)?
Which is cheaper per task, Terminus-2 – Llama 3.3 70B (Non-reasoning) or Codex – gpt-oss-20b (high)?
Which writes fewer output tokens per task, Terminus-2 – Llama 3.3 70B (Non-reasoning) or Codex – gpt-oss-20b (high)?
Which has the larger context window, Terminus-2 – Llama 3.3 70B (Non-reasoning) or Codex – gpt-oss-20b (high)?
More: Terminus-2 – Llama 3.3 70B (Non-reasoning) · Codex – gpt-oss-20b (high) · all comparisons