959M- Context
- 131Ktokens
- Released
- 1y ago
- Size
- 959Mparameters
Compact vision-language model for multilingual document parsing, OCR, tables, formulas, and charts
- Multimodal model
- Open weights · 959M
- Accepts images
How it compares
Additional source evidence
- Source model
- paddlepaddle/paddleocr-vl
- Last updated
- 2025-10-16
- Model context
- 131K
- Model output limit
- 16K
- Open weights
- Yes
- File attachments
- Yes
- Reasoning
- No
- Tool calling
- No
- Temperature control
- Yes
Downloadable weights
Complete source record (JSON)
{
"id": "paddlepaddle/paddleocr-vl",
"name": "PaddleOCR VL",
"description": "Compact vision-language model for multilingual document parsing, OCR, tables, formulas, and charts",
"last_updated": "2025-10-16",
"attachment": true,
"reasoning": false,
"tool_call": false,
"temperature": true,
"modalities": {
"input": [
"text",
"image"
],
"output": [
"text"
]
},
"limit": {
"context": 131072,
"output": 16384
},
"release_date": "2025-10-16",
"open_weights": true,
"license": "Apache-2.0",
"weights": [
{
"label": "Hugging Face",
"url": "https://huggingface.co/PaddlePaddle/PaddleOCR-VL"
}
]
}Frequently asked questions
Running locally
An estimated 1.2 GB at Q4, 1.7 GB at Q8 and 2.7 GB at FP16, based on its 959M parameters. All VRAM figures are estimates, covering the weights plus about 15% and 0.5 GB for a short context.
Yes. PaddleOCR VL fits on a 24 GB, 16 GB, 12 GB or 8 GB GPU at FP16, with an estimated 2.7 GB.
Yes. At Q4 the estimated VRAM is 1.2 GB, and assuming macOS lets the GPU use about 75% of unified memory, the smallest common Mac memory size that fits is 8 GB.
Privacy and data
Yes. PaddleOCR VL has open weights, so you can self-host it for full control of your data.
Capabilities and specs
The context window is 131K tokens (131,072), and the maximum output is 16K tokens (16,384).
Images yes. Its input modalities are text and image, so audio and video are not supported.
No. Tool calling (function calling) is not supported.
No. A reasoning or thinking mode is not supported.
Quality and benchmarks
Matched on model kind and release date with no comparable price, the closest alternatives are PaddleOCR-VL by Baidu, olmOCR 2 by Ai2 and GPT OSS Safeguard 120B by OpenAI.
PaddleOCR-VLPaddleOCR VL vs PaddleOCR-VLolmOCR 2PaddleOCR VL vs olmOCR 2GPT OSS Safeguard 120BPaddleOCR VL vs GPT OSS Safeguard 120B
Licensing and access
PaddleOCR VL is open weights, released under the Apache-2.0 license.
Yes. The license is Apache 2.0, a permissive license that allows commercial use.
The weights can be downloaded from the Hugging Face repository PaddlePaddle/PaddleOCR-VL.
Basics
PaddleOCR VL was made by Paddlepaddle and released on October 16, 2025.
Yes. PaddleOCR VL has open weights.
Generated from Vertical Tabs data, updated November 4, 2026.
Similar models
959M
8.3BolmOCR 2
GPT OSS Safeguard 120BGPT OSS Safeguard 120B
8BLlama 3.1 Nemotron Safety Guard 8B v3
12BNemotron Nano 12B v2 VL
Granite-4.0-H-Micro
Granite-4.0-H-Small
235BQwen3 VL 235B A22B Thinking
Gemma-SEA-LION-v4-27B-ITGemma-SEA-LION-v4-27B-IT
29.7BERNIE 4.5 VL 28B A3B Thinking
24BMagistral Small 1.2
229B
Discussion
No comments yet
Be the first to share your thoughts.