GOESB — Global Open Edge Speech Benchmark · pronounced “goose-bee”
Leaderboard
Objective, reproducible benchmarks for local (edge) speech AI. Every result here re-verified its hash and signature before being accepted — see why you can trust the numbers. New here? Read the docs.
This is every verified result, filterable and sortable. Looking for a single best recommendation for your constraints instead? Try the homepage →
AnyEvery verified result, batch and streaming mixed together.
Sorted by most recently submitted · 25 results · click a metric below to sort by it
| Profile | Runtime / Model | Hardware | Pack | Metrics | Submitted | Details |
|---|---|---|---|---|---|---|
| whisper-small-de-batch | faster-whisper / whisper-smallcuda | nvidia-rtx-5090 | fleurs-de | 2026-08-01T14:01:17Z | details
| |
| whisper-small-en-batch | faster-whisper / whisper-smallcuda | nvidia-rtx-5090 | librispeech-en | 2026-08-01T14:01:27Z | details
| |
| whisper-small-es-batch | faster-whisper / whisper-smallcuda | nvidia-rtx-5090 | fleurs-es | 2026-08-01T14:01:38Z | details
| |
| whisper-small-fr-batch | faster-whisper / whisper-smallcuda | nvidia-rtx-5090 | fleurs-fr | 2026-08-01T14:01:49Z | details
| |
| whisper-small-nl-batch | faster-whisper / whisper-smallcuda | nvidia-rtx-5090 | fleurs-nl | 2026-08-01T14:02:00Z | details
| |
| whisper-small-pt-batch | faster-whisper / whisper-smallcuda | nvidia-rtx-5090 | fleurs-pt | 2026-08-01T14:02:11Z | details
| |
| whisper-tiny-de-batch | faster-whisper / whisper-tinycuda | nvidia-rtx-5090 | fleurs-de | 2026-08-01T14:02:21Z | details
| |
| whisper-tiny-en-batch | faster-whisper / whisper-tinycuda | nvidia-rtx-5090 | librispeech-en | 2026-08-01T14:02:28Z | details
| |
| whisper-tiny-es-batch | faster-whisper / whisper-tinycuda | nvidia-rtx-5090 | fleurs-es | 2026-08-01T14:02:37Z | details
| |
| whisper-tiny-fr-batch | faster-whisper / whisper-tinycuda | nvidia-rtx-5090 | fleurs-fr | 2026-08-01T14:02:45Z | details
| |
| whisper-tiny-nl-batch | faster-whisper / whisper-tinycuda | nvidia-rtx-5090 | fleurs-nl | 2026-08-01T14:02:53Z | details
| |
| whisper-tiny-pt-batch | faster-whisper / whisper-tinycuda | nvidia-rtx-5090 | fleurs-pt | 2026-08-01T14:03:01Z | details
| |
| whispercpp-base-de-batch | whisper-cpp / whisper-base | amd-epyc-9135 | fleurs-de | 2026-08-01T14:04:02Z | details
| |
| whispercpp-base-en-batch | whisper-cpp / whisper-base.en | amd-epyc-9135 | librispeech-en | 2026-08-01T14:04:26Z | details
| |
| whispercpp-base-es-batch | whisper-cpp / whisper-base | amd-epyc-9135 | fleurs-es | 2026-08-01T14:04:45Z | details
| |
| whispercpp-base-fr-batch | whisper-cpp / whisper-base | amd-epyc-9135 | fleurs-fr | 2026-08-01T14:05:05Z | details
| |
| whispercpp-base-nl-batch | whisper-cpp / whisper-base | amd-epyc-9135 | fleurs-nl | 2026-08-01T14:05:24Z | details
| |
| whispercpp-base-pt-batch | whisper-cpp / whisper-base | amd-epyc-9135 | fleurs-pt | 2026-08-01T14:05:44Z | details
| |
| whispercpp-medium-de-batch | whisper-cpp / whisper-medium | amd-epyc-9135 | fleurs-de | 2026-08-01T14:11:36Z | details
| |
| whispercpp-medium-en-batch | whisper-cpp / whisper-medium.en | amd-epyc-9135 | librispeech-en | 2026-08-01T14:17:39Z | details
| |
| whispercpp-medium-es-batch | whisper-cpp / whisper-medium | amd-epyc-9135 | fleurs-es | 2026-08-01T14:20:53Z | details
| |
| whispercpp-medium-fr-batch | whisper-cpp / whisper-medium | amd-epyc-9135 | fleurs-fr | 2026-08-01T14:24:13Z | details
| |
| whispercpp-medium-nl-batch | whisper-cpp / whisper-medium | amd-epyc-9135 | fleurs-nl | 2026-08-01T14:27:31Z | details
| |
| whispercpp-medium-pt-batch | whisper-cpp / whisper-medium | amd-epyc-9135 | fleurs-pt | 2026-08-01T14:30:55Z | details
| |
| whispercpp-small-de-batch | whisper-cpp / whisper-small | amd-epyc-9135 | fleurs-de | 2026-08-01T14:32:24Z | details
|
Constant across this view: beam_size 5 · quantization int8 · threads 4 · vad true
* profile default