nvidia · audio asr

canary-1b-flash

nvidia/canary-1b-flash

canary-1b-flash at Q4_K_M is exactly 677,141,280 bytes (0.63 GiB / 0.68 GB) — an effective 6.679 bits per weight, not the nominal 4.

From the file· summed from 1 file(s)From the file· KV per layer
Parameters
811M
Architecture
canary
null layers
Context
native (config.json)
License
cc-by-4.0

Shipped quantizations

exact bytes, summed from published files
QuantSizeExact bytesEffective bpwTensorsPublisher
Q4_K_M0.63 GiB677,141,2806.679handy-computer
Q5_K_M0.72 GiB769,563,4247.591handy-computer
Q6_K0.80 GiB857,603,8728.459handy-computer
Q8_00.98 GiB1,048,131,36010.338handy-computer
F161.66 GiB1,785,657,12017.613handy-computer
F323.32 GiB3,560,372,000handy-computer

Measured

published by a third party, attributed below
MetricValueWhat it means
rtf2126.1
RTFx2126.1Higher is better — audio seconds processed per second of compute.
Word error rate5.38%Lower is better — the share of words transcribed incorrectly.
Word error rate8.26%Lower is better — the share of words transcribed incorrectly.
Word error rate12.11%Lower is better — the share of words transcribed incorrectly.
Word error rate10.58%Lower is better — the share of words transcribed incorrectly.
Word error rate12.03%Lower is better — the share of words transcribed incorrectly.
Word error rate8.19%Lower is better — the share of words transcribed incorrectly.
Word error rate1.20%Lower is better — the share of words transcribed incorrectly.
Word error rate2.52%Lower is better — the share of words transcribed incorrectly.
Word error rate1.75%Lower is better — the share of words transcribed incorrectly.
Word error rate4.17%Lower is better — the share of words transcribed incorrectly.
Word error rate5.78%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

Will it run on your card?

full quant x context sweep

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q4_K_M at roughly 0.42 GiB. The real file is 0.63 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

from config.json
Layers
Attention heads
KV heads
Head dim
Hidden size
Vocab
1,024
Sliding window
none
SWA period
MLA
no
Experts
Experts per token
use_sliding_window

Questions people ask

How much VRAM does canary-1b-flash need?
Q4_K_M is exactly 677,141,280 bytes (0.63 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-flash 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.