Terminus-2 – Kimi K3 (always on) vs. Terminus-2 – Llama 3.3 70B (Non-reasoning): geospatial agent comparison
Comparison between Terminus-2 – Kimi K3 (always on) and Terminus-2 – Llama 3.3 70B (Non-reasoning) 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 – Kimi K3 (always on) | Terminus-2 – Llama 3.3 70B (Non-reasoning) |
|---|---|---|
| Geospatial Agent Index | 58* | 7* |
| Earth-Bench V2 score | 61 | 7 |
| GeoBenchX score | 55 | 6 |
| GeoAgentBench score (guided GIS workflow; outside the index) | 94 | 8 |
| GeoAnalystBench score (guided GIS workflow; outside the index) | 89 | 0 |
| Raster and remote sensing score | 59 | 4 |
| Vector and overlay score | 33 | 0 |
| Networks and routing score | No data | No data |
| Climate and time series score | 66 | 10 |
| Spatial statistics and interpolation score | 50 | 0 |
| Mapping and cartography score | 78 | 2 |
| Recognising infeasible requests score | 38 | 13 |
| Cost per task | $0.336 | $0.024 |
| Time per task | 3.9 min | 3.9 min |
| Tokens per attempt | 76k | 65k |
| Turns per attempt | 7.7 | 14.9 |
| Attempts decided | 320 | 320 |
| Input price per 1M tokens | $3.00 | $0.29 |
| Output price per 1M tokens | $15.00 | $2.25 |
| Context window | 1,000,000 | 24,000 |
| Image input | Yes | No |
| Developer | Moonshot AI | Meta |
Charts
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 – Kimi K3 (always on) | Moonshot AI | 58* | No pending judgements; coverage incomplete; 1 trace review declined by the judge | 2 of 2 | 320 of 614 planned attempts |
| Terminus-2 – Llama 3.3 70B (Non-reasoning) | Meta | 7* | No pending judgements; coverage incomplete | 2 of 2 | 320 of 614 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.
Data table: Cost per task
| Model | Creator | Cost per task | Coverage (attempts) |
|---|---|---|---|
| Terminus-2 – Kimi K3 (always on) | Moonshot AI | $0.336 | 320 of 614 planned attempts |
| Terminus-2 – Llama 3.3 70B (Non-reasoning) | Meta | $0.024 | 320 of 614 planned attempts |
Time per task
Average agent wall time per task · Lower is better
First eight selected entries in metric order. Full results in table view.
Data table: Time per task
| Model | Creator | Time per task | Coverage (attempts) |
|---|---|---|---|
| Terminus-2 – Kimi K3 (always on) | Moonshot AI | 3.9 min | 320 of 614 planned attempts |
| Terminus-2 – Llama 3.3 70B (Non-reasoning) | Meta | 3.9 min | 320 of 614 planned attempts |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, Terminus-2 – Kimi K3 (always on) or Terminus-2 – Llama 3.3 70B (Non-reasoning)?
Terminus-2 – Kimi K3 (always on), at 58 against 7 for Terminus-2 – Llama 3.3 70B (Non-reasoning). Terminus-2 – Kimi K3 (always on): 320 of 614 planned attempts. Terminus-2 – Llama 3.3 70B (Non-reasoning): 320 of 614 planned attempts. Coverage is partial; scores can change as more attempts are decided.
Which is cheaper per task, Terminus-2 – Kimi K3 (always on) or Terminus-2 – Llama 3.3 70B (Non-reasoning)?
Terminus-2 – Llama 3.3 70B (Non-reasoning), at $0.024 against $0.336 for Terminus-2 – Kimi K3 (always on).
Which is faster per task, Terminus-2 – Kimi K3 (always on) or Terminus-2 – Llama 3.3 70B (Non-reasoning)?
Terminus-2 – Llama 3.3 70B (Non-reasoning), at 3.9 min against 3.9 min for Terminus-2 – Kimi K3 (always on).
Which has the larger context window, Terminus-2 – Kimi K3 (always on) or Terminus-2 – Llama 3.3 70B (Non-reasoning)?
Terminus-2 – Kimi K3 (always on), at 1,000,000 tokens against 24,000 tokens for Terminus-2 – Llama 3.3 70B (Non-reasoning).
More: Terminus-2 – Kimi K3 (always on) · Terminus-2 – Llama 3.3 70B (Non-reasoning) · all comparisons