Mistral Large 4 (high) vs Muse Glimmer 30B (high)
Results updated 6 November 2026 · Both in Terminus-2 · How we benchmark
Verdict
Muse Glimmer 30B (high) edges Mistral Large 4 (high) by 1 point, and is cheaper and faster.
- Cost per task
- $0.10 vs $0.078
- Time per task
- 11.4 min vs 3.0 min
- Tokens per attempt
- 233k vs 343k
- Image input
- Yes vs Yes
Mistral Large 4 (high), Terminus-2 scores higher on GS-QA2; Muse Glimmer 30B (high), Terminus-2 scores higher on Earth-Bench V2 and GeoBenchX. Muse Glimmer 30B (high), Terminus-2 is cheaper and faster: $0.078 against $0.10 per task, and 3.0 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) | Muse Glimmer 30B (high) |
|---|---|---|
| Spatial statistics | 46 | 8 |
| Recognising infeasible requests | 32 | 67 |
| Climate and time series | 43 | 61 |
| GIS analysis | 85 | 69 |
Where the tasks are set: each kind of work links to its tasks and scoresCharts for each: Mistral Large 4 (high), Terminus-2 · Muse Glimmer 30B (high), Terminus-2.
Side by side
| Metric | Mistral Large 4 (high), Terminus-2 | Muse Glimmer 30B (high), Terminus-2 |
|---|---|---|
| Geospatial Agent Index: Text only | 63 | 64 |
| Geospatial Agent Index: Text + vision | No data | No data |
| Earth-Bench V2 score | 50 | 60 |
| GeoBenchX score | 59 | 61 |
| GS-QA2 score | 79 | 70 |
| Vision agents score | No data | No data |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 80 | 79 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 94 | 58 |
| OGD4All score (standalone; in no index) | 95 | 95 |
| GeoNatureAgent score (standalone; in no index) | 100 | 100 |
| MultiGlobeQA score (standalone; in no index) | 71 | 72 |
| 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 | 63 |
| GIS analysis score | 85 | 69 |
| Climate and time series score | 43 | 61 |
| Spatial statistics score | 46 | 8 |
| Making maps score | 84 | 75 |
| Recognising infeasible requests score | 32 | 67 |
| Cost per task | $0.10 | $0.078 |
| Time per task (at 100 output tokens/s) | 11.4 min | 3.0 min |
| Tokens per attempt | 233k | 343k |
| Turns per attempt | 14.3 | 25.3 |
| Input price per 1M tokens | $0.70 | $0.35 |
| Output price per 1M tokens | $2.10 | $1.50 |
| Context window | 262,144 | 131,072 |
| Image input | Yes | Yes |
| Developer | Mistral AI | Meta |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, Mistral Large 4 (high), Terminus-2 or Muse Glimmer 30B (high), Terminus-2?
Muse Glimmer 30B (high), Terminus-2, 64 to 63.
Which is cheaper per task, Mistral Large 4 (high), Terminus-2 or Muse Glimmer 30B (high), Terminus-2?
Muse Glimmer 30B (high), Terminus-2, $0.078 to $0.10.
Which writes fewer output tokens per task, Mistral Large 4 (high), Terminus-2 or Muse Glimmer 30B (high), Terminus-2?
Muse Glimmer 30B (high), Terminus-2, 11k to 14k.
Which has the larger context window, Mistral Large 4 (high), Terminus-2 or Muse Glimmer 30B (high), Terminus-2?
Mistral Large 4 (high), Terminus-2, 262,144 tokens to 131,072 tokens.