nvidia · audio asr

canary-1b

nvidia/canary-1b

canary-1b at Q4_K_M is exactly 729,686,848 bytes (0.68 GiB / 0.73 GB) — an effective 5.715 bits per weight, not the nominal 4.

From the file· summed from 1 file(s)
Parameters
1.0B
Architecture
canary
Context
—
native (config.json)
License
cc-by-nc-4.0

Shipped quantizations

● exact bytes, summed from published files
QuantSize●Exact bytes●Effective bpw●Tensors●Publisher
Q4_K_M0.68 GiB729,686,8485.715—handy-computer
Q5_K_M0.78 GiB837,694,2726.561—handy-computer
Q6_K0.87 GiB934,167,6167.316—handy-computer
Q8_01.08 GiB1,162,740,2889.106—handy-computer
F161.91 GiB2,047,537,72816.036—handy-computer
F323.81 GiB4,086,100,54432.002—handy-computer

Measured

published by a third party, attributed below
MetricValueWhat it means
rtf765.6
RTFx765.6Higher is better — audio seconds processed per second of compute.
Word error rate1.86%Lower is better — the share of words transcribed incorrectly.
Word error rate12.42%Lower is better — the share of words transcribed incorrectly.
Word error rate1.15%Lower is better — the share of words transcribed incorrectly.
Word error rate11.63%Lower is better — the share of words transcribed incorrectly.
Word error rate2.53%Lower is better — the share of words transcribed incorrectly.
Word error rate5.52%Lower is better — the share of words transcribed incorrectly.
Word error rate4.01%Lower is better — the share of words transcribed incorrectly.
Word error rate8.47%Lower is better — the share of words transcribed incorrectly.
Word error rate8.55%Lower is better — the share of words transcribed incorrectly.
Word error rate10.87%Lower is better — the share of words transcribed incorrectly.
Word error rate5.79%Lower is better — the share of words transcribed incorrectly.
Benchmarked· by open-asr-leaderboard-english-short-latest

RTFx measured by the Open ASR Leaderboard on a single datacenter GPU at a large batch size. It ranks models against each other; it says nothing about throughput on consumer hardware. We reproduce these figures with attribution; they are not ours and we have not verified the runs. Source: open-asr-leaderboard-english-short-latest.

Compare with

same modality, comparable size
vs nemotron-3.5-asr-streaming-0.6bvs parakeet-unified-en-0.6bvs cohere-transcribe-03-2026vs parakeet-tdt-0.6b-v3vs whisper-mediumvs whisper-large-v3All alternatives →Will it fit on my card? →

Will it run on your card?

full quant x context sweep
By VRAM:8GB12GB16GB24GB32GB48GB
Radeon RX 6500 XT 4GBGeForce RTX 3050 6GBGeForce RTX 5050 8GBGeForce RTX 3080 10GBGeForce RTX 2080 Ti 11GBGeForce RTX 5070 12GBGeForce RTX 5060 Ti 16GBApple M3 Pro 18GBGeForce RTX 3080 Ti 20GBGeForce RTX 5090 D V2 24GBGeForce RTX 5090 D 32GBApple M5 Max 36GBApple M5 Pro 48GBApple M5 Max 64GBApple M3 Ultra 96GBApple M5 Max 128GBApple M2 Ultra 192GBApple M3 Ultra 256GBApple M3 Ultra 512GB

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q4_K_M at roughly 0.54 GiB. The real file is 0.68 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

Architecture unavailable — this repository is gated and no ungated mirror was found. Exact file sizes above are still authoritative; only the KV math needs the config.

Questions people ask

How much VRAM does canary-1b need?
Q4_K_M is exactly 729,686,848 bytes (0.68 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
Which quantization of canary-1b should I use?
Q4_K_M is the usual default. Pick the largest quantization that fits your card at the context you actually need — the table above gives exact sizes for every one published.