Models / NVIDIA/ Canary 1b Flash

Canary 1b Flash

NVIDIA · nvidia/canary-1b-flash

Speech to textTranscribes a recording into words

Input: audio. Output: text.InputOutput
Type
Open weightsCreative Commons Attribution 4.0
Languages
4
Size
1B

Context measured in tokens

Our take

Written Aug 2, 2026

Canary 1b Flash is NVIDIA's tiny downloadable speech-to-text model that turns audio into written words. It is extremely fast and near-perfect on clean recordings, but accuracy drops sharply on messy real-world audio and no hosts currently offer it.

Who should pick it

Pick this for lightning-fast batch transcription of clean audio, where it processes an hour of speech in about two seconds. Use it for financial call transcription or lightweight local deployment. Skip it if your audio is messy, accented, or from recorded meetings, where error rates rise tenfold; or if you need hosted inference.

The case for it

  • Processes audio at over two-thousand-times real time — an hour of audio in roughly two seconds on benchmark hardware.
  • Near-perfect on clean read-aloud audio at 1.2% word error rate, and strong on financial calls at 1.75%.
  • Creative Commons Attribution 4.0 licence allows commercial use with attribution.
  • European-accented speech is handled at 4.17% error, roughly three times better than its overall accented-speech rate.

The case against it

  • Accuracy collapses on messy real-world audio: recorded meetings hit 10.58% error, podcasts and video 8.19%, and accented speech overall 12.03%.
  • No hosted inference options are currently tracked.
  • Only four languages are listed, and all accuracy figures are English-only; German, Spanish and French are unverified.
00

How good is it?

TranscriptionTurning speech into text3 of 5Open ASR WER · 39th of 74

Words it gets right

94.2%

Misses roughly one word in 17, averaged over nine English test sets.

How fast it listens

2,126×18th of 62

an hour of audio in 2 seconds, on the board's own hardware. Your machine will differ.

Languages

4

Listed on the model card. The accuracy above is English only.

Where it struggles
Read aloudaudiobooks, clean recording1.2%
Podcasts and videoeveryday internet audio8.2%
Accented speechspeakers from many countries12%
Meetingsa room, several people, far microphone10.6%

Percentage of words wrong on each set, lower better. Bars are scaled to this model's own worst case, not to the board.

Which languages

English · German · Spanish · French

The figures above come from the Open ASR Leaderboard, an independent public test that runs every model on the same recordings. It is the only measurement of transcription quality we know of, so there are no other scores to show.

Also scored, on boards we give no mark for
Financial calls 8th of 74Harder read speech 17th of 74Clean read speech 21st of 74Podcasts and video 35th of 74Recorded meetings 37th of 74European-accented speech 49th of 74Accented speech 51st of 74

These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.

Every published score for this model9 scoresEvery figure we hold, from 9 boards, with who ran it and a link to the source — including the boards no rating above is built on.
2126.1independentsource ↗
5.8independentsource ↗
12independentsource ↗
1.8independentsource ↗
10.6independentsource ↗
8.2independentsource ↗
1.2independentsource ↗
2.5independentsource ↗
01

Can you run it yourself?

Fits in memory
weights load entirely on the card
Spills to system RAM
some weights offload; much slower
Too large
will not load even with offload
est
size is calculated; the verdict could change by 10%

Comfortable fit

A card many people ownFits in memory

GeForce RTX 4090 · 24 GB

Weights at Q4_K_M0.6 / 24 GBest
Spare memory20.9 GB spare
Usable context262Kwhat the spare memory holds; no published limit on record
Decode speed1330 tok/sest

Room to spare. 20.9 GB spare means a 10% error in the size would not change the answer.

One step upFits in memory

Apple M1 Pro (16-core GPU) · 32 GB

Weights at Q4_K_M0.6 / 32 GBest
Spare memory22.1 GB spare
Usable context262Kwhat the spare memory holds; no published limit on record
Decode speed231 tok/sest

Room to spare. 22.1 GB spare means a 10% error in the size would not change the answer.

On a MacFits in memory

Apple M2 (8-core GPU, 8GB unified) · 8 GB

Weights at Q4_K_M0.6 / 8 GBest
Spare memory4.1 GB spare
Usable context131Kwhat the spare memory holds; no published limit on record
Decode speed116 tok/sest

Room to spare. 4.1 GB spare means a 10% error in the size would not change the answer.

Your hardware
Checking your profile…

Memory use by level

Against a 24 GB card.

Q4_K_M
recommended
0.6 GBest
Fits in memory
Q5_K_M
0.7 GBest
Fits in memory
Q8_0
1.1 GBest
Fits in memory

All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.

Check against your own machine → · All 71 devices, with every size →

02

Models people weigh against Canary 1b Flash

03

When we formed this view

Dates behind this page

Aug 2, 2026BenchmarkScored 12 on Accented speechleaderboard
Aug 2, 2026BenchmarkScored 1.8 on Financial callsleaderboard
Aug 2, 2026BenchmarkScored 10.6 on Recorded meetingsleaderboard
Aug 2, 2026BenchmarkScored 4.2 on European-accented speechleaderboard
Aug 2, 2026BenchmarkScored 8.2 on Podcasts and videoleaderboard
Aug 2, 2026BenchmarkScored 1.2 on Clean read speechleaderboard
Aug 2, 2026BenchmarkScored 2.5 on Harder read speechleaderboard
Aug 1, 2026ListedListed on LLMapfirst indexed by our pipeline
Aug 1, 2026BenchmarkScored 2126.1 on Open ASR RTFxleaderboard
Aug 1, 2026BenchmarkScored 5.8 on Open ASR WERleaderboard

Prices last checked 6h ago

What we do not know about this model yet

  • We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
  • Nothing we hold says whether an endpoint streams, so we do not show it either way.
04

Licence and identifiers

What the licence allowsCreative Commons Attribution 4.0, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.

Licence

Creative Commons Attribution 4.0

permissiveCommercial use allowed

Permissive content license: any use with attribution. Common for datasets and some model weights.

Identifiers

Modality record
audio->text
Catalogue slug
nvidia-canary-1b-flash

Machine-readable model card (omc.json) →

Something wrong on this page? Tell us