Mistral Large 4 (high) vs MiniMax M3 (Reasoning)
Results updated 5 November 2026 · In Terminus-2 and MiniMax Code, so part of any gap can come from the harness · How we benchmark
Verdict
MiniMax M3 (Reasoning) beats Mistral Large 4 (high) by 5 points, and is faster; Mistral Large 4 (high) is cheaper.
- Cost per task
- $0.10 vs $0.15
- Time per task
- 11.4 min vs 4.9 min (estimated)
- Tokens per attempt
- 233k vs 1.68M
- Image input
- Yes vs Yes
Mistral Large 4 (high), Terminus-2 scores higher on GS-QA2; MiniMax M3 (Reasoning), MiniMax Code scores higher on Earth-Bench V2; they are level on GeoBenchX. Mistral Large 4 (high), Terminus-2 is cheaper ($0.10 against $0.15 per task), but MiniMax M3 (Reasoning), MiniMax Code is faster (4.9 min against 11.4 min).
Where they differ most
By kind of work, share of tasks passed, 0 to 100 · Higher is better
| Kind of work | Mistral Large 4 (high) | MiniMax M3 (Reasoning) |
|---|---|---|
| Climate and time series | 43 | 74 |
| Spatial statistics | 46 | 62 |
| Making maps | 84 | 77 |
| GIS analysis | 85 | 81 |
Where the tasks are set: each kind of work links to its tasks and scoresCharts for each: Mistral Large 4 (high), Terminus-2 · MiniMax M3 (Reasoning), MiniMax Code.
Side by side
| Metric | Mistral Large 4 (high), Terminus-2 | MiniMax M3 (Reasoning), MiniMax Code |
|---|---|---|
| Geospatial Agent Index: Text only | 63 | 68 |
| Geospatial Agent Index: Text + vision | No data | No data |
| Earth-Bench V2 score | 50 | 68 |
| GeoBenchX score | 59 | 59 |
| GS-QA2 score | 79 | 78 |
| Vision agents score | No data | No data |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 80 | 94 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 94 | 79 |
| OGD4All score (standalone; in no index) | 95 | 94 |
| GeoNatureAgent score (standalone; in no index) | 100 | 93 |
| MultiGlobeQA score (standalone; in no index) | 71 | 73 |
| GeoAgentEval score (standalone; in no index) | 92 | 92 |
| ImpactMesh score (standalone; in no index) | No data | No data |
| Remote sensing and raster analysis score | 69 | 73 |
| GIS analysis score | 85 | 81 |
| Climate and time series score | 43 | 74 |
| Spatial statistics score | 46 | 62 |
| Making maps score | 84 | 77 |
| Recognising infeasible requests score | 32 | 33 |
| Cost per task | $0.10 | $0.15 |
| Time per task (at 100 output tokens/s) | 11.4 min | 4.9 min (estimated) |
| Tokens per attempt | 233k | 1.68M |
| Turns per attempt | 14.3 | No data |
| Input price per 1M tokens | $0.70 | $0.30 |
| Output price per 1M tokens | $2.10 | $1.20 |
| Context window | 262,144 | 1,048,576 |
| Image input | Yes | Yes |
| Developer | Mistral AI | MiniMax |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, Mistral Large 4 (high), Terminus-2 or MiniMax M3 (Reasoning), MiniMax Code?
MiniMax M3 (Reasoning), MiniMax Code, 68 to 63.
Which is cheaper per task, Mistral Large 4 (high), Terminus-2 or MiniMax M3 (Reasoning), MiniMax Code?
Mistral Large 4 (high), Terminus-2, $0.10 to $0.15.
Which writes fewer output tokens per task, Mistral Large 4 (high), Terminus-2 or MiniMax M3 (Reasoning), MiniMax Code?
Mistral Large 4 (high), Terminus-2, 14k to 29k.
Which has the larger context window, Mistral Large 4 (high), Terminus-2 or MiniMax M3 (Reasoning), MiniMax Code?
MiniMax M3 (Reasoning), MiniMax Code, 1,048,576 tokens to 262,144 tokens.