|
|
--- |
|
|
license: apache-2.0 |
|
|
--- |
|
|
|
|
|
# Model Card: InternVL3_5-1B GGUF (Q4_0 Quantized) |
|
|
|
|
|
## Quick Overview |
|
|
A compact 1.1B parameter multimodal AI model that understands both images and text. Quantized to 4-bit for efficiency. |
|
|
|
|
|
**Key Features:** |
|
|
- **Size:** ~484 MB (compressed from original 1.1B parameters) |
|
|
- **Capabilities:** Image description, visual QA, document understanding, basic reasoning |
|
|
- **Format:** GGUF (runs on CPU/GPU) |
|
|
- **License:** Apache 2.0 |
|
|
|
|
|
## What It Does |
|
|
- Describe images and answer questions about them |
|
|
- Read text in images (OCR) |
|
|
- Understand documents and charts |
|
|
- Basic multimodal reasoning |
|
|
|
|
|
## Quick Start |
|
|
Use with: |
|
|
- LM Studio |
|
|
- llama.cpp |
|
|
- Ollama |
|
|
- Any GGUF-compatible software |
|
|
|
|
|
## Files |
|
|
- `InternVL3_5-1B-Q4_0.gguf` - Main model file |
|
|
- `mmproj-InternVL3_5-1B-Q4_0.gguf` - Multimodal projection file |
|
|
|
|
|
## Limitations |
|
|
- Not for medical or safety-critical use |
|
|
- May make mistakes (verify important outputs) |
|
|
- Reduced precision due to compression |
|
|
|
|
|
## Original Model |
|
|
- **Developer:** OpenGVLab |
|
|
- **Full Name:** InternVL3.5-1B |
|
|
- **Training:** Advanced multimodal training with reinforcement learning |
|
|
|
|
|
## Repository Contents |
|
|
The repository contains the following files: |
|
|
1. `.gitattributes` |
|
|
2. `InternVL3_5-1B-Q4_0.gguf` |
|
|
3. `README.md` (this file) |
|
|
4. `mmproj-InternVL3_5-1B-Q4_0.gguf` |
|
|
|
