GOESB

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 →

Lowest energyFastest (lowest RTF)Most accurate (lowest WER)
AnyEvery verified result, batch and streaming mixed together.
AnyBatchStreamingConcurrency

Sorted by most recently submitted · 7 results · click a metric below to sort by it

ProfileRuntime / ModelHardwarePackbeam_sizeMetricsSubmittedDetails
whisper-large-v3-de-batchfaster-whisper / whisper-large-v3intel-n150fleurs-de5 *
cer: 0.0054 ratiower: 0.0160 ratioram_mb: 4574.64 MBcpu_pct: 370.87 %temperature_c: 75 °Creal_time_factor: 2.04 ratio
2026-07-29T07:27:19Z
details
Profile
whisper-large-v3-de-batch@1.1.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cpu
beam_size
5
quantization
int8
threads
4
vad
true
whisper-medium-de-batchfaster-whisper / whisper-mediumintel-n150fleurs-de5 *
cer: 0.0073 ratiower: 0.0280 ratioram_mb: 2523.48 MBcpu_pct: 369.46 %temperature_c: 76.50 °Creal_time_factor: 1.09 ratio
2026-07-29T07:46:23Z
details
Profile
whisper-medium-de-batch@1.1.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cpu
beam_size
5
quantization
int8
threads
4
vad
true
whisper-small-de-batchfaster-whisper / whisper-smallintel-n150fleurs-de5 *
cer: 0.0133 ratiower: 0.0400 ratioram_mb: 958.86 MBcpu_pct: 360.48 %temperature_c: 77.50 °Creal_time_factor: 0.3876 ratio
2026-07-29T07:55:51Z
details
Profile
whisper-small-de-batch@1.1.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cpu
beam_size
5
quantization
int8
threads
4
vad
true
whispercpp-medium-de-batchwhisper-cpp / whisper-mediumintel-n150fleurs-de—
cer: 0.0066 ratiower: 0.0240 ratioram_mb: 1868.61 MBcpu_pct: 399.25 %temperature_c: 77.50 °Creal_time_factor: 4.87 ratio
2026-07-29T08:32:09Z
details
Profile
whispercpp-medium-de-batch@1.1.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cpu
threads
4
whispercpp-small-de-batchwhisper-cpp / whisper-smallintel-n150fleurs-de—
cer: 0.0181 ratiower: 0.0440 ratioram_mb: 725.74 MBcpu_pct: 398.92 %temperature_c: 75.50 °Creal_time_factor: 1.43 ratio
2026-07-29T09:07:43Z
details
Profile
whispercpp-small-de-batch@1.1.0
Pack
fleurs-de@1.0.0
Repeats
2
Backend
cpu
threads
4
whisper-large-v3-nl-batchfaster-whisper / whisper-large-v3intel-n150fleurs-nl2
cer: 0.0100 ratiower: 0.0438 ratioram_mb: 4662.71 MBcpu_pct: 370.63 %temperature_c: 75 °Creal_time_factor: 2.87 ratio
2026-07-29T06:11:21Z
details
Profile
whisper-large-v3-nl-batch@1.1.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
cpu
beam_size
2 (default 5)
quantization
int8
threads
4
vad
true
whisper-large-v3-nl-batchfaster-whisper / whisper-large-v3intel-n150fleurs-nl5 *
cer: 0.0095 ratiower: 0.0401 ratioram_mb: 4726.80 MBcpu_pct: 369.12 %temperature_c: 75 °Creal_time_factor: 3.18 ratio
2026-07-28T21:44:01Z
details
Profile
whisper-large-v3-nl-batch@1.1.0
Pack
fleurs-nl@1.0.0
Repeats
2
Backend
cpu
beam_size
5
quantization
int8
threads
4
vad
true

Constant across this view: quantization int8 · threads 4 · vad true

* profile default