---
license: other
license_name: webbrain-noncommercial-research-restrictions
base_model: webbrain-one/webbrain-compass-vld-tiny-v5
pipeline_tag: image-text-to-text
library_name: gguf
tags:
- webbrain
- minicpm-v
- gguf
- llama.cpp
- browser-agent
- decision-model
- system-one
- webbrain-compass
- webbrain-vld
---

# WebBrain Compass VLD Tiny v5 GGUF

GGUF exports of [`webbrain-one/webbrain-compass-vld-tiny-v5`](https://huggingface.co/webbrain-one/webbrain-compass-vld-tiny-v5) (`openbmb/MiniCPM-V-4.7-1B` fine-tune) configured with `qwen35.decision.type = "webbrain-vld"` and embedded `systemone` template so it runs in `llama.cpp` **both as a multimodal LLM (`/v1/chat/completions`) and as a multipass decision model (`/v1/systemone`)**.

## Files

| File | Quantization | Size |
|---|---|---:|
| `webbrain-compass-vld-tiny-v5-BF16.gguf` | `BF16` | `1.9 GB` |
| `webbrain-compass-vld-tiny-v5-Q8_0.gguf` | `Q8_0` | `1.1 GB` |
| `webbrain-compass-vld-tiny-v5-Q4_K_M.gguf` | `Q4_K_M` | `641 MB` |
| `mmproj-webbrain-compass-vld-tiny-v5-F16.gguf` | Vision Projector (`F16`) | `1.1 GB` |

## Usage with `llama-server`

```bash
llama-server \
  -m webbrain-compass-vld-tiny-v5-Q8_0.gguf \
  --mmproj mmproj-webbrain-compass-vld-tiny-v5-F16.gguf \
  -ngl 99 -c 16384 --port 8080
```