Capability breakdown
Intelligence over time
Provider & region options
Fastest loop: Overshoot · US East (Virginia) · 92 ms end to end.
| Overshoot | US East (Virginia) | 423 tok/s | 35 ms | 92 ms | $0.12 | <200 ms |
| Overshoot | US West (Oregon) | 423 tok/s | 45 ms | 102 ms | $0.12 | <200 ms |
| Overshoot | EU West (Ireland) | 423 tok/s | 51 ms | 107 ms | $0.12 | <200 ms |
| Overshoot | EU Central (Frankfurt) | 423 tok/s | 56 ms | 113 ms | $0.12 | <200 ms |
| Overshoot | Asia Pacific (Tokyo) | 423 tok/s | 73 ms | 130 ms | $0.12 | <200 ms |
| Fireworks | US East (Virginia) | 299 tok/s | 51 ms | 131 ms | $0.12 | <200 ms |
| Overshoot | Asia Pacific (Mumbai) | 423 tok/s | 81 ms | 138 ms | $0.12 | <200 ms |
| Overshoot | Middle East (Dubai) | 423 tok/s | 84 ms | 141 ms | $0.12 | <200 ms |
| Fireworks | US West (Oregon) | 299 tok/s | 62 ms | 142 ms | $0.12 | <200 ms |
| Overshoot | South America (São Paulo) | 423 tok/s | 93 ms | 149 ms | $0.12 | <200 ms |
| Together | US East (Virginia) | 262 tok/s | 58 ms | 150 ms | $0.12 | <200 ms |
| Fireworks | EU West (Ireland) | 299 tok/s | 74 ms | 154 ms | $0.12 | <200 ms |
| Together | US West (Oregon) | 262 tok/s | 73 ms | 164 ms | $0.12 | <200 ms |
| Together | EU West (Ireland) | 262 tok/s | 74 ms | 166 ms | $0.12 | <200 ms |
| vLLM (self-host) | US East (Virginia) | 212 tok/s | 74 ms | 188 ms | $0.08 | <200 ms |
| vLLM (self-host) | EU West (Ireland) | 212 tok/s | 83 ms | 197 ms | $0.08 | <200 ms |
Sample outputs
Illustrative samples shared across models, not verbatim output from Holo3 35B A3B.
Scene description
“Describe what's happening in this camera frame.”
A forklift is reversing toward loading bay 3 while a worker in a hi-vis vest signals from the left. Two pallets remain unwrapped near the dock edge.
OCR + extraction
“Read the shipping label and return structured JSON.”
{ "tracking": "1Z A24 9W7 03 8421 990 4", "carrier": "UPS", "weight_kg": 12.4, "dest": "Portland, OR 97204" }
Grounding
“Point to every person not wearing a hard hat.”
2 detections, [x:0.41, y:0.62, w:0.08, h:0.19] conf 0.94 · [x:0.77, y:0.55, w:0.07, h:0.21] conf 0.88.
Where it ranks
- #8Gemini 3 Flash75.3
- #9Claude Haiku 4.574.2
- #10Holo3 35B A3B74.0
- #11Gemma 4 31B73.0
- #12Gemma 4 26B A4B69.3
- #13GPT-5.4 nano63.6