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.
modelFun-ASR-MLT-Nano-2512
AMD 4750U GPU8.8×
AMD 4750U CPU5.5×
Q8_0 size0.83 GB
frontier · AMD 4750U GPU8 of 31
verdict: Best in 2 of 31 languages, and tied with the leader in 3 more (overlapping 95% CIs). Weakest relative to the field in Polish (12.7× the leader's error), Greek (11.6×), and Romanian (10.8×). Faster than realtime on the AMD 4750U GPU (8.8×) and AMD 4750U CPU (5.5×).
license: FunASR Model Open Source License Agreement v1.1 · card languages: 32 · —
standing by language · 31 benchmarked
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.
Fun-ASR-MLT-Nano-2512 errorbest other model errortie: the 95% CIs overlap
quantization · 8 usable languages at Q5_K_M
Expect about 4% more errors at Q5_K_M, measurably worse on 2 of 8 usable languages, for 71% of the Q8_0 file.
Q5_K_M file71% of Q8_00.59 GB against 0.83 GB
median error change+3.8%over 8 usable languages
measurably worse2 of 8above the Q8_0 95% confidence interval