Kimi K2.6 (Reasoning), Terminus-2 vs. MiniMax M3.1 Flash (max), Terminus-2
Kimi K2.6 (Reasoning), Terminus-2 against MiniMax M3.1 Flash (max), Terminus-2: Geospatial Agent Index, each evaluation, cost, time and tokens, on the same tasks and limits. Methodology.
Side by side
| Metric | Kimi K2.6 (Reasoning), Terminus-2 | MiniMax M3.1 Flash (max), Terminus-2 |
|---|---|---|
| Geospatial Agent Index: Text only | 70 | 51 |
| Geospatial Agent Index: Text + vision | No data | No data |
| Earth-Bench V2 score | 68 | 39 |
| GeoBenchX score | 64 | 42 |
| GS-QA2 score | 77 | 71 |
| Vision agents score | No data | No data |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 90 | 88 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 79 | 47 |
| OGD4All score (standalone; in no index) | No data | No data |
| GeoNatureAgent score (standalone; in no index) | No data | No data |
| MultiGlobeQA score (standalone; in no index) | No data | No data |
| GeoAgentEval score (standalone; in no index) | No data | No data |
| Remote sensing and raster analysis score | 77 | 57 |
| GIS analysis score | 74 | 77 |
| Climate and time series score | 69 | 34 |
| Spatial statistics score | 15 | 25 |
| Making maps score | 77 | 57 |
| Recognising infeasible requests score | 57 | 30 |
| Cost per task | $0.089 | $0.017 |
| Time per task (at 100 output tokens/s) | 3.5 min | 8.4 min |
| Tokens per attempt | 175k | 254k |
| Turns per attempt | 12.8 | 13.2 |
| Input price per 1M tokens | $0.95 | $0.15 |
| Output price per 1M tokens | $4.00 | $0.60 |
| Context window | 262,144 | 1,000,000 |
| Image input | Yes | Yes |
| Developer | Moonshot AI | Other |
Charts
Geospatial Agent Index
Equal average of Earth-Bench V2, GeoBenchX and GS-QA2, 0 to 100 · Higher is better
Data table: Geospatial Agent Index
| Model | Developer | Geospatial Agent Index |
|---|---|---|
| Kimi K2.6 (Reasoning), Terminus-2 | Moonshot AI | 70 |
| MiniMax M3.1 Flash (max), Terminus-2 | Other | 51 |
Cost per task
Average cost per task (USD) · Lower is better
Data table: Cost per task
| Model | Developer | Cost per task |
|---|---|---|
| MiniMax M3.1 Flash (max), Terminus-2 | Other | $0.017 |
| Kimi K2.6 (Reasoning), Terminus-2 | Moonshot AI | $0.089 |
Output tokens per task
Average output tokens per attempt, reasoning included · Lower is better
Data table: Output tokens per task
| Model | Developer | Output tokens per task |
|---|---|---|
| Kimi K2.6 (Reasoning), Terminus-2 | Moonshot AI | 12k |
| MiniMax M3.1 Flash (max), Terminus-2 | Other | 18k |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, Kimi K2.6 (Reasoning), Terminus-2 or MiniMax M3.1 Flash (max), Terminus-2?
Kimi K2.6 (Reasoning), Terminus-2, 70 to 51.
Which is cheaper per task, Kimi K2.6 (Reasoning), Terminus-2 or MiniMax M3.1 Flash (max), Terminus-2?
MiniMax M3.1 Flash (max), Terminus-2, $0.017 to $0.089.
Which writes fewer output tokens per task, Kimi K2.6 (Reasoning), Terminus-2 or MiniMax M3.1 Flash (max), Terminus-2?
Kimi K2.6 (Reasoning), Terminus-2, 12k to 18k.
Which has the larger context window, Kimi K2.6 (Reasoning), Terminus-2 or MiniMax M3.1 Flash (max), Terminus-2?
MiniMax M3.1 Flash (max), Terminus-2, 1,000,000 tokens to 262,144 tokens.
More: Kimi K2.6 (Reasoning), Terminus-2 · MiniMax M3.1 Flash (max), Terminus-2 · All comparisons