RK182X-VLM-MinerU2.5-Pro-2604-1.2B
1. Model Overview
This repository provides an RKNN VLM document parsing model converted from OpenDataLab/MinerU2.5-Pro-2604-1.2B. The official model is a 1.2B-parameter image-text-to-text model for parsing document layouts, text, formulas, and tables. The official weights are licensed under Apache License 2.0.Hugging Face model page.
- Model ID:
RKNNAI/RK182X-VLM-MinerU2.5-Pro-2604-1.2B - Model Display Name:
RK182X-VLM-MinerU2.5-Pro-2604-1.2B - RKNN Runtime Version:
v1.1.0 - Source Model:
OpenDataLab/MinerU2.5-Pro-2604-1.2B - Model Type: VLM
- Chip Label: RK182X
- Supported Chips: RK1820, RK1828
Available Models
| RKNN Runtime Version | Configuration Directory | Supported Chips | Resolution | Quantization | NPU Cores | Context Length (tokens) | KVCache |
|---|---|---|---|---|---|---|---|
v1.1.0 | MinerU2.5-Pro-2604-1.2B-1036x1036-w4a16-8-8192-normal | RK1828 | Vision: 1036x1036 | Vision: w4a16; LLM: w4a16 | Vision: 8; LLM: 8 | LLM: 8192 (8k) | LLM: fp16 |
v1.1.0 | MinerU2.5-Pro-2604-1.2B-448x448-w4a16-8-8192-normal | RK1820, RK1828 | Vision: 448x448 | Vision: w4a16; LLM: w4a16 | Vision: 8; LLM: 8 | LLM: 8192 (8k) | LLM: fp16 |
2. Files
| File or Directory | Description |
|---|---|
LICENSE | Source model license |
NOTICE | License, usage requirements, and attribution |
<configuration-directory>/ | Matching model files |
Subdirectory README.md | Configuration documentation (English) |
Subdirectory README_CN.md | Configuration documentation (Chinese) |
Subdirectory config.json | Model configuration and file manifest |
Subdirectory SHA256SUMS | SHA-256 checksums of the files in this directory (excluding this file) |
3. Download
Download the full repository from ModelScope:
modelscope download --model RKNNAI/RK182X-VLM-MinerU2.5-Pro-2604-1.2B --revision v1.1.0 --local_dir ./RK182X-VLM-MinerU2.5-Pro-2604-1.2B
Download the full repository from Hugging Face:
hf download RKNNAI/RK182X-VLM-MinerU2.5-Pro-2604-1.2B --revision v1.1.0 --local-dir ./RK182X-VLM-MinerU2.5-Pro-2604-1.2B
Download a specific configuration from ModelScope:
from modelscope import snapshot_download
snapshot_download(
"RKNNAI/RK182X-VLM-MinerU2.5-Pro-2604-1.2B",
revision="v1.1.0",
allow_patterns=["README.md", "README_CN.md", "LICENSE", "NOTICE", "MinerU2.5-Pro-2604-1.2B-1036x1036-w4a16-8-8192-normal/**"],
local_dir="./RK182X-VLM-MinerU2.5-Pro-2604-1.2B",
)
Download a specific configuration from Hugging Face:
hf download RKNNAI/RK182X-VLM-MinerU2.5-Pro-2604-1.2B --revision v1.1.0 --include "README.md" "README_CN.md" "LICENSE" "NOTICE" "MinerU2.5-Pro-2604-1.2B-1036x1036-w4a16-8-8192-normal/**" --local-dir ./RK182X-VLM-MinerU2.5-Pro-2604-1.2B
4. SHA-256 Verification
Run in the configuration directory:
cd ./RK182X-VLM-MinerU2.5-Pro-2604-1.2B/MinerU2.5-Pro-2604-1.2B-1036x1036-w4a16-8-8192-normal
sha256sum -c SHA256SUMS
Deploy only after all entries show OK.
5. Compatibility and Limitations
- See the Available Models table for the chips supported by each configuration.
- Use the matching RKNN Runtime and driver. Check the device-side version with
rknn-smi -vbefore running. - Use the files from the selected configuration directory together.
- Provides single-image document recognition; the application must implement the complete PDF parsing pipeline.
6. Copyright and License
The source model is licensed under Apache License 2.0. See LICENSE for the full text and NOTICE for provenance and conversion details. The applicable RKNN Toolkit and RKNN Runtime licenses also apply.