These numbers come from installs that opted in to anonymous usage reporting. Every field is listed on the telemetry page. We publish them back because the community generated them.
Covering 2026-07-18 to 2026-08-17. Self-reported by participating installs, so treat them as a good signal rather than a census.
| Active herds | 1 |
| Machines in those herds | 1 |
| Requests routed | 15,183 |
| Prompt tokens | 21,482,199 |
| Completion tokens | 3,758,320 |
Most-used models
| # | Model | Requests | Share |
|---|---|---|---|
| 1 | gpt-oss:120b | 15,181 | 99.99% |
| 2 | mlx:mlx-community/GLM-5-4bit | 1 | 0.01% |
| 3 | gemma3:27b | 1 | 0.01% |
Named herds
A few people have given their herd a name. Ranked by total memory, because that is more fun to argue about than request counts.
| # | Herd | Machines | Total memory |
|---|---|---|---|
| 1 | LiveNeon Geeks in the Woods | 1 | 512 GB |
Most common errors
| # | Category | Count | Share |
|---|---|---|---|
| 1 | unknown | 1 | 100.0% |
Error categories are a fixed list. Raw error text never leaves anyone's machine.
Machine-readable
The same data is available as JSON at /api/v1/stats.