Prompt: (raw) (yaml)
words:27 bytes:174
| model | words | bytes | total duration |
load duration |
prompt eval count |
prompt eval duration |
prompt eval rate |
eval count |
eval duration |
eval rate |
|---|---|---|---|---|---|---|---|---|---|---|
| granite3.2-vision:2b | 183 | 1134 | 29.986237333s | 12.217167ms | 2274 token(s) | 19.383799459s | 117.31 tokens/s | 242 token(s) | 10.588779166s | 22.85 tokens/s |
| minicpm-v:8b | 105 | 677 | 17.078933209s | 19.470042ms | 107 token(s) | 6.817980125s | 15.69 tokens/s | 196 token(s) | 10.240018917s | 19.14 tokens/s |
| qwen2.5vl:3b | 23 | 214 | 5.140854s | 23.936417ms | 144 token(s) | 2.994221875s | 48.09 tokens/s | 76 token(s) | 2.121731167s | 35.82 tokens/s |
| qwen2.5vl:7b | 26 | 166 | 7.286911042s | 24.449209ms | 144 token(s) | 2.726106167s | 52.82 tokens/s | 81 token(s) | 4.535051166s | 17.86 tokens/s |
| gemma3:4b | 37 | 321 | 47.868590125s | 30.051208ms | 306 token(s) | 42.305483041s | 7.23 tokens/s | 152 token(s) | 5.53211575s | 27.48 tokens/s |
| llava:7b | 109 | 770 | 24.040301917s | 12.843584ms | 627 token(s) | 4.149775s | 151.09 tokens/s | 348 token(s) | 19.875526167s | 17.51 tokens/s |
| llava-llama3:8b | 4 | 25 | 6.070579458s | 21.411292ms | 624 token(s) | 4.785659125s | 130.39 tokens/s | 21 token(s) | 1.261941s | 16.64 tokens/s |
| llava-phi3:3.8b | 48 | 298 | 10.281514917s | 11.089584ms | 629 token(s) | 2.952202584s | 213.06 tokens/s | 195 token(s) | 7.316052125s | 26.65 tokens/s |
| bakllava:7b | 4 | 26 | 5.565268084s | 9.610834ms | 631 token(s) | 4.225901s | 149.32 tokens/s | 25 token(s) | 1.328225542s | 18.82 tokens/s |
| moondream:1.8b | 4 | 28 | 5.667217333s | 16.158ms | 777 token(s) | 5.340658167s | 145.49 tokens/s | 19 token(s) | 308.658875ms | 61.56 tokens/s |
| System | |
| ollama proc | 100% GPU |
| ollama version | 0.9.0 |
| sys arch | arm64 |
| sys processor | arm |
| sys memory | 15G + 1216M |
| sys OS | Darwin 24.5.0 |