Model reference · open weights
PP-OCRv6_medium_rec_onnx is an open-weight language model from PaddlePaddle, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
PP-OCRv6: From 1.5M to 34.5M Parameters, Surpassing Billion-Scale VLMs on OCR Tasks [](https://github.com/PaddlePaddle/PaddleOCR) [](https://huggingface.co/PaddlePaddle/PP-OCRv6mediumreconnx) [](https://x.com/PaddlePaddle) [](./LICENSE) [](https://huggingface.co/PaddlePaddle/PP-OCRv6mediumrec) [](https://huggingface.co/PaddlePaddle/PP-OCRv6mediumrecsafetensors) 🔥 Official Website 📝 Technical Report PP-OCRv6 Overview PP-OCRv6 is a lightweight OCR system that combines architectural innovation with data-centric optimization. It redesigns the backbone, detection neck, and recognition neck around a unified MetaFormer-style building block with structural reparameterization. Three model tiers (medium, small, tiny) share the same block primitives, covering deployment scenarios from server to edge. Key Features 1. Unified and Scalable Model Family: A three-tier OCR model family spanning 1.5M to 34.5M parameters. PP-OCRv6medium achieves 86.2% detection Hmean and 83.2% recognition accuracy, outperforming PP-OCRv5server by +4.6% and +5.1% respectively. 2. Lightweight Architectural Innovations: (i) LCNetV4, a MetaFormer-style lightweight backbone with structural reparameterization; (ii) RepLKFPN, a detection neck with dilated reparameterizable depthwise convolutions; (iii) EncoderWithLightSVTR, a recognition neck with local-global attention and additive skip connections. 3. Multi-Language and Scenario Support: Supports 50 languages and diverse industrial scenes (digital displays, dot-matrix characters, tire prints, etc.), surpassing Qwen3-VL-235B, GPT-5.5, and Gemini-3.1-Pro with orders of magnitude fewer parameters. PP-OCRv6mediumrec Introduction PP-OCRv6 text recognition architecture overview PP-OCRv6mediumrec is the largest recognition model in the PP-OCRv6 series. It uses LCNetV4 as the backbone and EncoderWithLightSVTR as the recognition neck, with a CTC+NRTR multi-head decoder. The model supports 50 languages and contains 19M parameters. The key accuracy metrics are as follows: Quick Start Installation 1. PaddleOCR 2. ONNX Runtime environment (required for onnx models) Model Usage You can quickly experience the functionality with a single command: You can also integrat
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | PaddlePaddle |
|---|---|
| Type | Language models |
| Variants | 1 |
| Runs with | PaddleOCR |
| Released | 2026-06-09 |
| Popularity | 70k downloads / month |
| Likes | 24 |
| Licence | Open weights |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| PP-OCRv6_medium_rec_onnx | — | BF16 | — | — | Weights ↗ |
Using it via the API
Once AxForge deploys pp-ocrv6-medium-rec-onnx for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (pp-ocrv6-medium-rec-onnx below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/chat/completions \
-H "Authorization: Bearer $AXFORGE_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"pp-ocrv6-medium-rec-onnx","messages":[{"role":"user","content":"Hello"}]}'
Licence
Open weights under apache-2.0 — commercial use is permitted. Deploy it on AxForge EU hardware on request. Read the licence ↗