OCR models / acvlab

ABot-OCR

acvlab · released 2026-05-26 · data checked 2026-11-06

ABot-OCR is a 2.1B-parameter OCR and document-parsing model from acvlab, released 2026-05-26 under the Not stated license. It scores 93.30 on OmniDocBench v1.6 (#19 of 33). It needs about 5.3 GB of memory at 16-bit.

OmniDocBench v1.6

93.30

#19 of 33 models

olmOCR-Bench

—

Not reported

Parameters

2.1B

OCR specialist

Memory to run

~5.3 GB

16-bit weights plus about 1 GB

Benchmarks

OmniDocBench v1.6 Leaderboard93.30

Best: OXR 1.0 97.03

The tick marks the best score on each benchmark. See how to read the ranking.

Specifications

Lab: acvlab
Type: OCR / document parsing specialist
Total parameters: 2.1B
License: Not stated
No license is stated on the model card.
Memory to run: ~5.3 GB at 16-bit
Released: 2026-05-26

Weights and usage on Hugging Face ↗

Closest alternatives

Logics-Parsing v2 (4.4B, 93.33 on OmniDocBench, Not stated)
FireRed-OCR (2.1B, 93.26 on OmniDocBench, Apache-2.0)
MinerU2.5 (2509) (1.2B, 93.04 on OmniDocBench, AGPL-3.0)
Ovis2.6 30B-A3B (31.4B, 93.70 on OmniDocBench, Apache-2.0)

Sources

Something wrong here? Report it.