model detail

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.

model whisper-large-v3
AMD 4750U GPU2.1×
AMD 4750U CPU0.6×
Q8_0 size1.55 GB
frontier · AMD 4750U GPU15 of 40

verdict: Best in 15 of 40 languages, and tied with the leader in 8 more (overlapping 95% CIs). Weakest relative to the field in Cantonese (CER) (3.6× the leader's error), Maltese (3.5×), and Hindi (2.2×). Fast enough for realtime on the AMD 4750U GPU (2.1×), not on the AMD 4750U CPU (0.6×).

license: Apache-2.0 · card languages: 100 · translatelang detecttimestamps (segment)

standing by language · 40 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.

whisper-large-v3 error best other model error tie: the 95% CIs overlap

speed · all 19 models on the AMD 4750U GPU

× realtime, log scale. Left of the realtime marker, transcribing takes longer than the recording. Hover a dot for the model.

compare with

whisper-large-v3-turbosame familywhisper-mediumsame familywhisper-smallsame familycanary-1b-v2ahead of it in 13 of 25 shared languagesFun-ASR-MLT-Nano-2512closest coverage, 30 languages in common