Benchmark run

Hviske vs main models

This benchmark compares the hviske model to the already main models recommended in Codictate

Aug 18, 2026

Models: 7
Samples per dataset: 200
Warmup runs: 3
Hardware: Apple M4 Max / 36 GB / macOS 26.5.1

Full results

Accuracy is the share of words transcribed correctly per language condition, so higher is better; speed is milliseconds per second of audio, so lower is better. The best value in each column is emphasised, and a cell reading a dash says on hover why there is no figure in it.

Accuracy (%) per language, and speed in ms per second of audio, so lower is better. A cell reading a dash was never measured, and hovering any cell says what is behind it.


Hviske v5 Tiny Q5_0	197	181 MB	282 MB	16 ms	88.7%Every word from every language counted together, not an average of per-language scores.	Not scored on any benchmark language.	88.7%Every word from every language counted together, not an average of per-language scores.	88.7%
Hviske v5 Tiny Q4_K	197	153 MB	253 MB	16 ms	88.6%Every word from every language counted together, not an average of per-language scores.	Not scored on any benchmark language.	88.6%Every word from every language counted together, not an average of per-language scores.	88.6%
Hviske v5 Tiny F16	197	503 MB	601 MB	21 ms	88.5%Every word from every language counted together, not an average of per-language scores.	Not scored on any benchmark language.	88.5%Every word from every language counted together, not an average of per-language scores.	88.5%
Hviske v5 Tiny Q8_0	197	268 MB	368 MB	18 ms	88.5%Every word from every language counted together, not an average of per-language scores.	Not scored on any benchmark language.	88.5%Every word from every language counted together, not an average of per-language scores.	88.5%
Hviske v5 Tiny Q6_K	197	232 MB	332 MB	17 ms	88.3%Every word from every language counted together, not an average of per-language scores.	Not scored on any benchmark language.	88.3%Every word from every language counted together, not an average of per-language scores.	88.3%
Whisper Large V3 Turbo Q5_0	197	574 MB	742 MB	52 ms	84.8%Every word from every language counted together, not an average of per-language scores.	Not scored on any benchmark language.	84.8%Every word from every language counted together, not an average of per-language scores.	84.8%
Parakeet TDT 0.6B v3	197	500 MB	79 MB	17 ms	80.6%Every word from every language counted together, not an average of per-language scores.	Not scored on any benchmark language.	80.6%Every word from every language counted together, not an average of per-language scores.	80.6%

At a glance

Ratings computed from benchmark data, scaled 1 to 10, and written with their scale in every cell. Accuracy is based on Word Error Rate (WER) and does not include punctuation yet.

Ratings run 1 to 10, higher is better. A cell reading a dash was never measured, and hovering any cell says what is behind it.


Hviske v5 Tiny F16	Danish only	No	9/10	9/10Every word from every language counted together, not an average of per-language scores.
Hviske v5 Tiny Q8_0	Danish only	No	10/10	9/10Every word from every language counted together, not an average of per-language scores.
Hviske v5 Tiny Q6_K	Danish only	No	10/10	9/10Every word from every language counted together, not an average of per-language scores.
Hviske v5 Tiny Q5_0	Danish only	No	10/10	9/10Every word from every language counted together, not an average of per-language scores.
Hviske v5 Tiny Q4_K	Danish only	No	10/10	9/10Every word from every language counted together, not an average of per-language scores.
Whisper Large V3 Turbo Q5_0	All languages	No	9/10	8/10Every word from every language counted together, not an average of per-language scores.
Parakeet TDT 0.6B v3	25 languages	No	10/10	7/10Every word from every language counted together, not an average of per-language scores.

Charts

The plots generated alongside this run. Select a chart to open it full size.

Bar chart comparing model accuracy across English, Spanish, Danish, and Hungarian benchmark conditions. — Accuracy by model and test condition

Bar chart comparing transcription speed across models and test conditions. — Speed comparison across conditions

Bar chart showing average English, multilingual, and overall accuracy per model. — Average accuracy by language group