Overview
In Terminus-2 it ranks 66th of 74 entries on the Geospatial Agent Index, at 33.8. It costs $0.071 per task, less than the median entry ($0.081) and takes 7.9 min per task, slower than the median entry (2.0 min).
Compare head to head: gpt-oss-20b (high), Terminus-2 · Mistral Large 3 (Non-reasoning), Terminus-2 · Gemini 3.1 Flash-Lite (minimal), Terminus-2 · gpt-oss-120b (low), Terminus-2 · GLM-4.7-Flash (Reasoning), Claude Code · gpt-oss-120b (high), Codex · all comparisons.
Technical specification
- Developer
- StepFun
- Weights
- Open
- Harness
- Terminus-2 2.0.0
- Reasoning setting
- Reasoning
- Context window
- 1,024,000 tokens
- Maximum output tokens
- 65,536
- Image input
- Yes
- Temperature
- 0.6
- Input price per 1M tokens
- $1.00
- Cached input price per 1M tokens
- $0.050
- Output price per 1M tokens
- $2.70
- Map and trace judge
- Claude Sonnet 5.5
- Reward hacking
- 1.9% of reviewed passing attempts flaggedSafety page
Agent tasks by evaluation
All
Geospatial Agent Index: Text only
Equal average of Earth-Bench V2, GeoBenchX and GS-QA2, 0 to 100 · Higher is better
Data table: Geospatial Agent Index: Text only
Open weights
Geospatial Agent Index: Text only
Equal average of Earth-Bench V2, GeoBenchX and GS-QA2, 0 to 100 · Higher is better
Data table: Geospatial Agent Index: Text only
All
Earth-Bench V2 score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Earth observation data analysis. Earth-Bench V2 results and background.
Data table: Earth-Bench V2 score
Open weights
Earth-Bench V2 score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Earth observation data analysis. Earth-Bench V2 results and background.
Data table: Earth-Bench V2 score
All
GeoBenchX score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Maps, numbers and lists from country and city data; some tasks should be declined. GeoBenchX results and background.
Data table: GeoBenchX score
Open weights
GeoBenchX score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Maps, numbers and lists from country and city data; some tasks should be declined. GeoBenchX results and background.
Data table: GeoBenchX score
All
GS-QA2 score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Spatial data reasoning over map and elevation data. GS-QA2 results and background.
Data table: GS-QA2 score
Open weights
GS-QA2 score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Spatial data reasoning over map and elevation data. GS-QA2 results and background.
Data table: GS-QA2 score
All
GeoAgentBench score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Multi-step spatial execution. GeoAgentBench results and background.
Data table: GeoAgentBench score
Open weights
GeoAgentBench score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Multi-step spatial execution. GeoAgentBench results and background.
Data table: GeoAgentBench score
All
GeoAnalystBench score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Spatial workflow and code generation. GeoAnalystBench results and background.
Data table: GeoAnalystBench score
Open weights
GeoAnalystBench score
Share of tasks passed, 0 to 100 · Higher is better
How to read this chart
About this evaluation. Spatial workflow and code generation. GeoAnalystBench results and background.
Data table: GeoAnalystBench score
Cost
All
Cost per task
Average cost per agent task (USD) · Lower is better
How to read this chart
What cost is measuring. What one attempt at a task would cost if paid token by token at the provider's published API prices, averaged over attempts: ordinary input tokens, input tokens served from the provider's cache at the lower cached price, tokens written to the cache where the provider charges for that, and output tokens (thinking included). Cost assumes each harness uses the model's pay-per-token API, priced at public list rates; sandboxes and our own engineering time are not included. Failed attempts count too.
How to read this chart. Lower is better. Costs are pay-per-token list prices, not any discount or credit a customer might have.
Data table: Cost per task
Open weights
Cost per task
Average cost per agent task (USD) · Lower is better
How to read this chart
What cost is measuring. What one attempt at a task would cost if paid token by token at the provider's published API prices, averaged over attempts: ordinary input tokens, input tokens served from the provider's cache at the lower cached price, tokens written to the cache where the provider charges for that, and output tokens (thinking included). Cost assumes each harness uses the model's pay-per-token API, priced at public list rates; sandboxes and our own engineering time are not included. Failed attempts count too.
How to read this chart. Lower is better. Costs are pay-per-token list prices, not any discount or credit a customer might have.
Data table: Cost per task
All
Geospatial Agent Index vs. cost per task
Geospatial Agent Index, Text only; cost per task (USD) · Higher and further left is better
- Most attractive quadrant
- Pareto line
How to read this chart
Why a log scale. Costs and times differ by a factor of ten or more between models, so equal steps on the axis are equal ratios.
Not yet ranked. Entries with too few tasks run so far are not plotted; the table lists them, and every entry is on the models page.
What cost is measuring. What one attempt at a task would cost if paid token by token at the provider's published API prices, averaged over attempts: ordinary input tokens, input tokens served from the provider's cache at the lower cached price, tokens written to the cache where the provider charges for that, and output tokens (thinking included). Cost assumes each harness uses the model's pay-per-token API, priced at public list rates; sandboxes and our own engineering time are not included. Failed attempts count too.
How to read this chart. Each point is one entry. Further left means a lower average cost per task (log scale); higher means a higher Geospatial Agent Index. Both axes fit the models shown, so the score axis does not start at zero. The shaded most attractive quadrant is the top-left quarter of the plot, split at the middle of each axis: a reading aid that moves with the axes when you change the models shown, not a fixed bar. The dotted Pareto line joins the shown models that no cheaper shown model beats on score.
Data table: Geospatial Agent Index vs. cost per task
Open weights
Geospatial Agent Index vs. cost per task
Geospatial Agent Index, Text only; cost per task (USD) · Higher and further left is better
- Most attractive quadrant
- Pareto line
How to read this chart
Why a log scale. Costs and times differ by a factor of ten or more between models, so equal steps on the axis are equal ratios.
What cost is measuring. What one attempt at a task would cost if paid token by token at the provider's published API prices, averaged over attempts: ordinary input tokens, input tokens served from the provider's cache at the lower cached price, tokens written to the cache where the provider charges for that, and output tokens (thinking included). Cost assumes each harness uses the model's pay-per-token API, priced at public list rates; sandboxes and our own engineering time are not included. Failed attempts count too.
How to read this chart. Each point is one entry. Further left means a lower average cost per task (log scale); higher means a higher Geospatial Agent Index. Both axes fit the models shown, so the score axis does not start at zero. The shaded most attractive quadrant is the top-left quarter of the plot, split at the middle of each axis: a reading aid that moves with the axes when you change the models shown, not a fixed bar. The dotted Pareto line joins the shown models that no cheaper shown model beats on score.
Data table: Geospatial Agent Index vs. cost per task
| Model | Geospatial Agent Index | Cost per task (USD) | On the Pareto line |
|---|---|---|---|
| DeepSeek V4 Pro (max), Terminus-2 | 71 | $0.35 | No |
| Gemma 4 26B A4B (Reasoning), Terminus-2 | 56 | $0.021 | Yes |
| gpt-oss-120b (high), Terminus-2 | 46 | $0.026 | No |
| gpt-oss-20b (high), Terminus-2 | 33 | $0.017 | No |
| Kimi K3 (Reasoning), Terminus-2 | 72 | $0.23 | Yes |
| Llama 3.3 70B (Non-reasoning), Terminus-2 | 8 | $0.023 | No |
| MiniMax M3.1 Flash (max), Terminus-2 | 51 | $0.017 | No |
| Mistral Large 3 (Non-reasoning), Terminus-2 | 33 | $0.026 | No |
| Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 | 20 | $0.13 | No |
| Nemotron 3 Ultra (high), Terminus-2 | 68 | $0.069 | No |
| Nemotron 3.5 Lightning 30B (Reasoning), Terminus-2 | 11 | $0.012 | No |
| Nemotron 3.5 Super VL 120B (Reasoning), Terminus-2 | 52 | $0.0096 | Yes |
| Step 5 Preview (Reasoning), Terminus-2 | 34 | $0.071 | No |
| GLM-5.3-Flash (max), Claude Code | 72 | $0.093 | Yes |
| GLM-5.3 (max), Claude Code | 73 | $0.48 | Yes |
| DeepSeek V4.1 Flash (max), Codex | 71 | $0.067 | Yes |
| Kimi K2.7 Code (Reasoning), Kimi Code | 72 | $0.18 | Yes |
| Qwen 3.8 27B (xhigh), Qwen Code | 68 | $0.22 | No |
Time
All
Time per task
Average minutes per agent task, at 100 output tokens/s · Lower is better
How to read this chart
What this metric means. Time as if every model wrote 100 output tokens per second: output tokens divided by 100, plus the time the agent's commands took. Serving speed differs between providers and is not ranked.
Data table: Time per task
Open weights
Time per task
Average minutes per agent task, at 100 output tokens/s · Lower is better
How to read this chart
What this metric means. Time as if every model wrote 100 output tokens per second: output tokens divided by 100, plus the time the agent's commands took. Serving speed differs between providers and is not ranked.
Data table: Time per task
All
Geospatial Agent Index vs. time per task
Geospatial Agent Index, Text only; time per task (minutes) · Higher and further left is better
- Most attractive quadrant
- Pareto line
How to read this chart
Why a log scale. Costs and times differ by a factor of ten or more between models, so equal steps on the axis are equal ratios.
What this metric means. Time as if every model wrote 100 output tokens per second: output tokens divided by 100, plus the time the agent's commands took. Serving speed differs between providers and is not ranked.
Not yet ranked. Entries with too few tasks run so far are not plotted; the table lists them, and every entry is on the models page.
How to read this chart. Each point is one entry. Further left means a lower average time per task (log scale); higher means a higher Geospatial Agent Index. Both axes fit the models shown, so the score axis does not start at zero. The shaded most attractive quadrant is the top-left quarter of the plot, split at the middle of each axis: a reading aid that moves with the axes when you change the models shown, not a fixed bar. The dotted Pareto line joins the shown models that no faster shown model beats on score.
Data table: Geospatial Agent Index vs. time per task
Open weights
Geospatial Agent Index vs. time per task
Geospatial Agent Index, Text only; time per task (minutes) · Higher and further left is better
- Most attractive quadrant
- Pareto line
How to read this chart
Why a log scale. Costs and times differ by a factor of ten or more between models, so equal steps on the axis are equal ratios.
What this metric means. Time as if every model wrote 100 output tokens per second: output tokens divided by 100, plus the time the agent's commands took. Serving speed differs between providers and is not ranked.
How to read this chart. Each point is one entry. Further left means a lower average time per task (log scale); higher means a higher Geospatial Agent Index. Both axes fit the models shown, so the score axis does not start at zero. The shaded most attractive quadrant is the top-left quarter of the plot, split at the middle of each axis: a reading aid that moves with the axes when you change the models shown, not a fixed bar. The dotted Pareto line joins the shown models that no faster shown model beats on score.
Data table: Geospatial Agent Index vs. time per task
| Model | Geospatial Agent Index | Time per task (minutes) | On the Pareto line |
|---|---|---|---|
| DeepSeek V4 Pro (max), Terminus-2 | 71 | 4.9 min | No |
| Gemma 4 26B A4B (Reasoning), Terminus-2 | 56 | 5.6 min | No |
| gpt-oss-120b (high), Terminus-2 | 46 | 3.0 min | No |
| gpt-oss-20b (high), Terminus-2 | 33 | 7.2 min | No |
| Kimi K3 (Reasoning), Terminus-2 | 72 | 2.7 min | Yes |
| Llama 3.3 70B (Non-reasoning), Terminus-2 | 8 | 2.1 min | No |
| MiniMax M3.1 Flash (max), Terminus-2 | 51 | 8.6 min | No |
| Mistral Large 3 (Non-reasoning), Terminus-2 | 33 | 1.1 min | Yes |
| Nemotron 3 Nano Omni 30B (Reasoning), Terminus-2 | 20 | 4.9 min | No |
| Nemotron 3 Ultra (high), Terminus-2 | 68 | 2.4 min | No |
| Nemotron 3.5 Lightning 30B (Reasoning), Terminus-2 | 11 | 4.9 min | No |
| Nemotron 3.5 Super VL 120B (Reasoning), Terminus-2 | 52 | 3.8 min | No |
| Step 5 Preview (Reasoning), Terminus-2 | 34 | 7.9 min | No |
| GLM-5.3-Flash (max), Claude Code | 72 | 6.5 min | No |
| GLM-5.3 (max), Claude Code | 73 | 4.3 min | Yes |
| DeepSeek V4.1 Flash (max), Codex | 71 | 8.6 min | No |
| Kimi K2.7 Code (Reasoning), Kimi Code | 72 | 1.9 min | Yes |
| Qwen 3.8 27B (xhigh), Qwen Code | 68 | 4.7 min | No |
Tokens
All
Output tokens per task
Average output tokens per agent task, reasoning included · Lower is better
How to read this chart
Output tokens per task. The average number of tokens the model wrote per attempt, its reasoning included, as reported by the provider.
Compare with care. Models use different tokenisers, so the same text is a different number of tokens for different models. Use token counts to see how much work each model did, and cost to compare spending.
Data table: Output tokens per task
Open weights
Output tokens per task
Average output tokens per agent task, reasoning included · Lower is better
How to read this chart
Output tokens per task. The average number of tokens the model wrote per attempt, its reasoning included, as reported by the provider.
Compare with care. Models use different tokenisers, so the same text is a different number of tokens for different models. Use token counts to see how much work each model did, and cost to compare spending.
Data table: Output tokens per task
All
Turns per task
Average agent turns per task · Lower is better
How to read this chart
What this metric means. A turn is one model reply plus the commands it ran.
Data table: Turns per task
Open weights
Turns per task
Average agent turns per task · Lower is better
How to read this chart
What this metric means. A turn is one model reply plus the commands it ran.