# MiniCPM-OCR-Jev

> MiniCPM-OCR Jev (2B) is an OCR quality-assessment model for text, formula, and table regions. Given a region image and a candidate OCR result, it returns a score from 0 to 1 — higher means the candidate is more likely correct.

- **Organisation:** openbmb
- **Category:** OCR & Document AI, Vision-Language Models
- **Parameters:** 2.2B
- **Context length:** 262,144 tokens
- **Licence:** Apache 2.0 (Commercial use OK)
- **Released:** 2026-10-23
- **Hugging Face:** 79.5K downloads (30 days), 100 likes

## Estimated memory

| Quant | ≈ GB |
|---|---|
| Q4_K_M | 2 |
| Q8_0 | 3.13 |
| F16 | 5.45 |

## Runs with

- [llama.cpp](https://selfhost.directory/project/llamacpp)
- [LM Studio](https://lmstudio.ai)
- [SGLang](https://selfhost.directory/project/sglang)
- [vLLM](https://selfhost.directory/project/vllm)
- [Jan](https://jan.ai) (community)
- [KoboldCpp](https://selfhost.directory/project/koboldcpp) (community)
- [LocalAI](https://selfhost.directory/project/localai) (community)
- [Ollama](https://selfhost.directory/project/ollama) (community)

## About

MiniCPM-OCR Jev (2B) is an OCR quality-assessment model for text, formula, and table regions. Given a region image and a candidate OCR result, it returns a score from 0 to 1 — higher means the candidate is more likely correct.

MiniCPM-OCR Jev is a 2B OCR quality-assessment model for text, formula, and table regions. Given a region image and a candidate OCR result, it returns a score from 0 to 1 — higher means the candidate is more likely correct. Use it to auto-accept reliable regions, route the rest to human review, or pick the best result among several OCR models.

• 🛡️ A small model that beats larger ones: at only 2B parameters, decision AUROC (×100) on
• JevOCRBench is 93.6, above
• JEV-35B-A3B (90.1) and close to Qwen3.5-397B-A17B (93.8).
• 🧩 Stable generalization: on parses from MiniCPM-OCR 1.0, GLM-OCR, PaddleOCR-VL-1.6, and
• Xiaomi-OCR-0, AUROC (×100) stays in 91.9–93.4.

Trained from Qwen3.5-2B. The vision encoder and language backbone are retained, and the language-model output head lm_head is replaced with a binary classification head that learns whether a candidate is correct.

When calling the cloud API, the Docker image, or the SDK service with a built-in renderer, supply the original region image, the candidate text, and the region category (text / formula / table) — the service renders internally. A service without a renderer requires you to supply the rendered image as well.

## Alternatives

- [baidu/Unlimited-OCR](https://selfhost.directory/models/baidu/unlimited-ocr) (3.3B (1.1B active), MIT, ≈4.46 GB): Please refer to the official vLLM recipe for deployment details:
- [baidu/Unlimited-OCR-2](https://selfhost.directory/models/baidu/unlimited-ocr-2) (3.3B (1.1B active), MIT, ≈4.47 GB): Sparse Reference Sliding Window Attention (SR-SWA): Compresses text-visual attention interactions from ultra-long sequences down to a fixed-size set of working references (e.g., 512 tokens).
- [deepseek-ai/DeepSeek-OCR](https://selfhost.directory/models/deepseek-ai/deepseek-ocr) (3.3B (1.1B active), MIT, ≈4.46 GB): Refer to 🌟GitHub for guidance on model inference acceleration and PDF processing, etc.
- [zai-org/GLM-OCR](https://selfhost.directory/models/zai-org/glm-ocr) (1.3B, MIT, ≈3.46 GB): GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture. It introduces Multi-Token Prediction (MTP) loss and stable full-task reinforcement learning to improve training efficiency, recognition accuracy, and generalization.
- [openbmb/MiniCPM-OCR-1.0](https://selfhost.directory/models/openbmb/minicpm-ocr-1.0) (1.3B, Apache 2.0, ≈3.41 GB): MiniCPM-OCR 1.0 (1.3B) is the first OCR-specific model in the MiniCPM-V/o family. It focuses on text-centric document understanding — text, tables, formulas, charts, seals, and structured information extraction — and balances quality against speed on real-world captures.
- [dots-studio/dots.mocr](https://selfhost.directory/models/dots-studio/dots.mocr) (3B, MIT, ≈4.11 GB): We present dots.mocr. Beyond achieving state-of-the-art (SOTA) performance in standard multilingual document parsing among models of comparable size, dots.mocr excels at converting structured graphics (e.g., charts, UI layouts, scientific figures and etc.) directly into SVG code.

More: https://selfhost.directory/models/openbmb/minicpm-ocr-jev/alternatives

## Runs well on

- [GeForce RTX 4060](https://selfhost.directory/models/openbmb/minicpm-ocr-jev/on/rtx-4060): F16
- [Apple M4 (16 GB)](https://selfhost.directory/models/openbmb/minicpm-ocr-jev/on/apple-m4-16gb): F16
- [Radeon RX 6700 XT](https://selfhost.directory/models/openbmb/minicpm-ocr-jev/on/rx-6700-xt): F16
- [GeForce RTX 3060 12GB](https://selfhost.directory/models/openbmb/minicpm-ocr-jev/on/rtx-3060): F16
- [Arc B580](https://selfhost.directory/models/openbmb/minicpm-ocr-jev/on/arc-b580): F16
- [Tesla T4](https://selfhost.directory/models/openbmb/minicpm-ocr-jev/on/tesla-t4): F16

## Similar size

- [nanonets/Nanonets-OCR2-1.5B-exp](https://selfhost.directory/models/nanonets/nanonets-ocr2-1.5b-exp) (1.6B)
- [opendatalab/MinerU-Diffusion-V1-0320-2.5B](https://selfhost.directory/models/opendatalab/mineru-diffusion-v1-0320-2.5b) (2.7B)

## FAQ

**How much VRAM does MiniCPM-OCR-Jev need?**
About 5.5 GB at FP16 / BF16.

**Can MiniCPM-OCR-Jev run on the RTX 4060?**
Yes. It fits at FP16 / BF16 (about 4.7 GB). It fits comfortably on the Apple M4 (16 GB), Radeon RX 6700 XT, RTX 3060 12GB.

**Can I use MiniCPM-OCR-Jev commercially?**
Yes. It is released under Apache 2.0. Permissive: use, modify and sell freely; keep the notice and licence text.

**How do I run MiniCPM-OCR-Jev with Ollama or llama.cpp?**
With Ollama: ollama run hf.co/openbmb/MiniCPM-OCR-Jev-gguf:Q4_K_M. With llama.cpp: llama-server -hf openbmb/MiniCPM-OCR-Jev-gguf:Q4_K_M, which starts an OpenAI-compatible server.

## Links

- huggingface: https://huggingface.co/openbmb/MiniCPM-OCR-Jev
- github: https://github.com/OpenBMB/MiniCPM-OCR

---
Source: https://selfhost.directory/models/openbmb/minicpm-ocr-jev