MiniMax M3.1 Flash (max), Terminus-2 vs. Claude Haiku 5.5 (low), Claude Code
MiniMax M3.1 Flash (max), Terminus-2 against Claude Haiku 5.5 (low), Claude Code: Geospatial Agent Index, each evaluation, cost, time and tokens, on the same tasks and limits. The harnesses differ, so part of any gap can come from the harness. Methodology.
Side by side
| Metric | MiniMax M3.1 Flash (max), Terminus-2 | Claude Haiku 5.5 (low), Claude Code |
|---|---|---|
| Geospatial Agent Index: Text only | 51 | 70 |
| Geospatial Agent Index: Text + vision | No data | 72 |
| Earth-Bench V2 score | 39 | 62 |
| GeoBenchX score | 41 | 73 |
| GS-QA2 score | 71 | 75 |
| Vision agents score | No data | 77 |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 88 | 92 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 47 | 100 |
| OGD4All score (standalone; in no index) | No data | 95 |
| GeoNatureAgent score (standalone; in no index) | No data | 100 |
| MultiGlobeQA score (standalone; in no index) | No data | 72 |
| GeoAgentEval score (standalone; in no index) | No data | 92 |
| Remote sensing and raster analysis score | 57 | 72 |
| GIS analysis score | 77 | 73 |
| Climate and time series score | 34 | 64 |
| Spatial statistics score | 22 | 38 |
| Making maps score | 53 | 79 |
| Recognising infeasible requests score | 33 | 73 |
| Cost per task | $0.017 | $0.0053 |
| Time per task (at 100 output tokens/s) | 8.6 min | 0.9 min |
| Tokens per attempt | 263k | 181k |
| Turns per attempt | 13.3 | 5.5 |
| Input price per 1M tokens | $0.15 | $0.10 |
| Output price per 1M tokens | $0.60 | $0.50 |
| Context window | 1,000,000 | 1,000,000 |
| Image input | Yes | Yes |
| Developer | MiniMax | Anthropic |
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 |
|---|---|---|
| Claude Haiku 5.5 (low), Claude Code | Anthropic | 70 |
| MiniMax M3.1 Flash (max), Terminus-2 | MiniMax | 51 |
Cost per task
Average cost per task (USD) · Lower is better
Data table: Cost per task
| Model | Developer | Cost per task |
|---|---|---|
| Claude Haiku 5.5 (low), Claude Code | Anthropic | $0.0053 |
| MiniMax M3.1 Flash (max), Terminus-2 | MiniMax | $0.017 |
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 |
|---|---|---|
| Claude Haiku 5.5 (low), Claude Code | Anthropic | 3k |
| MiniMax M3.1 Flash (max), Terminus-2 | MiniMax | 19k |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, MiniMax M3.1 Flash (max), Terminus-2 or Claude Haiku 5.5 (low), Claude Code?
Claude Haiku 5.5 (low), Claude Code, 70 to 51.
Which is cheaper per task, MiniMax M3.1 Flash (max), Terminus-2 or Claude Haiku 5.5 (low), Claude Code?
Claude Haiku 5.5 (low), Claude Code, $0.0053 to $0.017.
Which writes fewer output tokens per task, MiniMax M3.1 Flash (max), Terminus-2 or Claude Haiku 5.5 (low), Claude Code?
Claude Haiku 5.5 (low), Claude Code, 3k to 19k.
Which has the larger context window, MiniMax M3.1 Flash (max), Terminus-2 or Claude Haiku 5.5 (low), Claude Code?
They are level, at 1,000,000 tokens.
More: MiniMax M3.1 Flash (max), Terminus-2 · Claude Haiku 5.5 (low), Claude Code · All comparisons