---
title: d1-2B-GGUF
canonical_url: "https://www.modelscope.cn/models/LiquidAI/d1-2B-GGUF"
md_url: "https://www.modelscope.cn/models/LiquidAI/d1-2B-GGUF.md"
repository: LiquidAI/d1-2B-GGUF
last_updated: 2026-10-29
license: other
pipeline_tag: image-text-to-text
tasks:
- image-text-to-text
base_model:
- LiquidAI/d1-2B
base_model_relation: quantized
library_name:
- gguf
- pytorch
frameworks:
- pytorch
language:
- en
- ar
- zh
- fr
- de
- it
- ja
- ko
- pt
- es
- vi
- th
- id
- hi
- ru
- pl
downloads: 40
stars: 0
tags:
- liquid
- lfm3
- edge
- decision
- classification
- extraction
- calibration
- system-one
- multimodal
- decision-model
- gguf
- llama.cpp
---
# d1-2B-GGUF
> d1-2B-GGUF - LiquidAI 在 ModelScope 开源的模型。GGUF version of d1-2B for local inference on CPUs and GPUs.
LiquidAI/d1-2B-GGUF 是 ModelScope 魔搭社区上的image-text-to-text模型，采用 other 许可，基于 LiquidAI/d1-2B 构建。
- **Repository**: LiquidAI/d1-2B-GGUF
- **License**: other
- **Tasks**: image-text-to-text
- **Base model**: LiquidAI/d1-2B
- **Tags**: liquid, lfm3, edge, decision, classification, extraction, calibration, system-one, multimodal, decision-model, gguf, llama.cpp
- **Downloads**: 40
- **Stars**: 0
- **Last updated**: 2026-10-29
Source: https://www.modelscope.cn/models/LiquidAI/d1-2B-GGUF
---
# d1-2B-GGUF
GGUF version of [d1-2B](https://huggingface.co/LiquidAI/d1-2B) for local inference on CPUs and GPUs.
d1-2B is a decision model for text, pictures, and video, built on [LFM3-VL-2B](https://huggingface.co/LiquidAI/LFM3-VL-2B). It returns calibrated answers in one forward pass, without generating tokens.
Find architecture and usage details in the [original model card](https://huggingface.co/LiquidAI/d1-2B).
## Use with llama.cpp
> [!NOTE]
> Supported in [llama.cpp](https://github.com/ggml-org/llama.cpp) [b11960](https://github.com/ggml-org/llama.cpp/releases/tag/b11960) ([v0.8.0](https://github.com/ggml-org/llama.cpp/releases/tag/v0.8.0)) and newer.
llama.cpp answers `noul`, `choice` and `score` questions over a state (text or JSON), pictures or a video. Video input needs `ffmpeg` in `PATH`. Multiple choice (`"multiple": true`), extraction, estimates and boxes are not supported yet.
```bash
llama-server -hf LiquidAI/d1-2B-GGUF:Q8_0 -b 8192 -ub 8192
```
A question is read in one batch, so `-b` and `-ub` must hold its state and media. A picture takes at most half of `-ub`: 8192 keeps the full budget of 4096 tokens per picture, smaller values lower it.
Then send questions to the `/v1/systemone` endpoint.
### Text
```bash
curl http://127.0.0.1:8080/v1/systemone -H "Content-Type: application/json" -d '{
"state": "I was charged twice this month, please refund one of them.",
"questions": {
"refund": {"type": "noul", "instructions": "Is the customer asking for a refund?"},
"team": {"type": "choice", "instructions": "Which team should handle this?",
"criteria": {"billing": "Charges, refunds, invoices", "technical": "App or site faults",
"fraud": "Suspected unauthorised use"}},
"urgency": {"type": "score", "instructions": "How urgent is this?",
"criteria": ["Can wait", "Today", "Blocking the customer now"]}
}
}'
```
### Image + Text
```bash
curl -sL -o cats.jpg http://images.cocodataset.org/val2017/000000039769.jpg # two cats on a sofa
curl http://127.0.0.1:8080/v1/systemone -H "Content-Type: application/json" -d @- <
API
Try
Docs
Blog