HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

nemotron-ocr-v1

nvidia/nemotron-ocr-v1

kind
original
quant
bf16 (16-bit)
family
nemotron
modality
image-to-text
root model
nvidia/nemotron-ocr-v1
provider
nvidia
license
other
language
en
context
—
library
—
vRAM (GB)
—
size on disk
—
downloads
135
likes
121
created
—
last modified
2026-04-01T22:09:55+00:00 (2026)
huggingface
https://huggingface.co/nvidia/nemotron-ocr-v1
tags
imageocrobject recognitiontext recognitionlayout analysisingestionimage-to-textenlicense:otherregion:us

Benchmark scores (3)

benchmarkcategoryscorecontextconfevidence
CERspeech-asr16.33Character Error Ratemedium| Character Error Rate | 0.1633 | 0.2029 | -19.5% ✔️ |
WERspeech-asr12.03Bag-of-word Error Ratemedium| Bag-of-word Error Rate | 0.1203 | 0.2748 | -56.2% ✔️ |
CERspeech-asr4.53Bag-of-character Error Ratemedium| Bag-of-character Error Rate | 0.0453 | 0.0512 | -11.5% ✔️ |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.