---

license: openrail

language:

• en

base_model:

• datalab-to/lift

pipeline_tag: image-text-to-text

library_name: transformers

tags:

• text-generation-inference
• pdf
• extraction
• structured-data
• json
• llama-cpp

---

lift-GGUF

> lift is a 9-billion-parameter structured extraction model from Datalab, built on Qwen3.5-9B, that pulls structured JSON data out of PDFs and images by accepting any JSON Schema and returning a matching JSON object via schema-constrained decoding to guarantee valid, well-typed output. It handles multi-page documents in a single pass — including values spanning multiple pages — supports both local (HuggingFace) and remote (vLLM server) inference modes, and ships with a CLI for single files, inline schemas, or whole directories, plus a Streamlit-based Schema Studio for building and testing schemas. On a 225-document benchmark (6–64 pages, ~11,000 scored fields) featuring adversarial cases like cross-page values, exhaustive lists, and near-miss distractors, lift achieves 90.2% field accuracy and 20.9% full-document accuracy at a 9.5s median latency, outperforming similarly-sized open models like Qwen3.5-9B (76.3%) and NuExtract3 (81.5%) while trailing larger hosted systems like the Datalab API (95.9%) and Gemini Flash 3.5 (91.3%); code is Apache 2.0 licensed while model weights use a modified OpenRAIL-M license that's free for research, personal use, and startups under $5M in funding or revenue.

Model Files

File Name | Quant Type | File Size | File Link |

|-----------|------------|-----------|-----------|

| lift.BF16.gguf | BF16 | 18.4 GB | Download |

| lift.F16.gguf | F16 | 18.4 GB | Download |

| lift.Q2_K.gguf | Q2_K | 3.91 GB | Download |

| lift.Q3_K_L.gguf | Q3_K_L | 5.05 GB | Download |

| lift.Q3_K_M.gguf | Q3_K_M | 4.74 GB | Download |

| lift.Q3_K_S.gguf | Q3_K_S | 4.36 GB | Download |

| lift.Q4_0.gguf | Q4_0 | 5.45 GB | Download |

| lift.Q4_K_M.gguf | Q4_K_M | 5.78 GB | Download |

| lift.Q4_K_S.gguf | Q4_K_S | 5.49 GB | Download |

| lift.Q5_0.gguf | Q5_0 | 6.47 GB | Download |

| lift.Q5_K_M.gguf | Q5_K_M | 6.64 GB | Download |

| lift.Q5_K_S.gguf | Q5_K_S | 6.47 GB | Download |

| lift.Q6_K.gguf | Q6_K | 7.56 GB | Download |

| lift.Q8_0.gguf | Q8_0 | 9.79 GB | Download |

| lift.mmproj-bf16.gguf | mmproj-bf16 | 922 MB | Download |

| lift.mmproj-f16.gguf | mmproj-f16 | 922 MB | Download |

| lift.mmproj-q8_0.gguf | mmproj-q8_0 | 624 MB | Download |

llama.cpp

LLM inference in C/C++ — https://github.com/ggml-org/llama.cpp