Where this model stands against the best model in each language at Q8_0, its speed on an AMD Ryzen 4750U laptop, and what the card says. Add a second model to compare them head to head.
modelcanary-1b-v2
AMD 4750U GPU14.1×
AMD 4750U CPU8.3×
Q8_0 size1.07 GB
frontier · AMD 4750U GPU16 of 25
verdict: Best in 11 of 25 languages, and tied with the leader in 4 more (overlapping 95% CIs). Weakest relative to the field in Greek (2.9× the leader's error), Ukrainian (1.9×), and Russian (1.6×). Faster than realtime on the AMD 4750U GPU (14.1×) and AMD 4750U CPU (8.3×).
Each bar compares this model with the best other model in that language, on a log scale: left of center it is ahead, right of center behind, and the other model is named on the row. The two error rates use the legend colors: this model first, the other model second. Faded bars are ties, where the 95% confidence intervals overlap. Sorted strongest standing first. Hover a row for full details.
canary-1b-v2 errorbest other model errortie: the 95% CIs overlap
quantization · 25 usable languages at Q5_K_M
Q5_K_M is within noise on every one of 25 usable languages: 73% of the Q8_0 file for no measurable accuracy cost.
Q5_K_M file73% of Q8_00.78 GB against 1.07 GB
median error change+0.9%over 25 usable languages
measurably worse0 of 25above the Q8_0 95% confidence interval