language:
- en
- zh
base_model: LiquidAI/LFM3-VL-2B
pipeline_tag: image-text-to-text
tags: - lfm3_vl
- abliterix
- edge
license: other
license_name: lfm1.0
license_link: LICENSE
LFM3-VL-2B-Uncensored
A vision-language model based on LiquidAI/LFM3-VL-2B, modified with Abliterix to reduce refusal behavior, for local and edge-device deployment.
The model supports text and image inputs, retaining the vision encoder and multimodal projector while modifying the language backbone's attention weights with layer-specific refusal directions.
Model details
| Property | Description |
|---|---|
| Base model | LiquidAI/LFM3-VL-2B |
| Task | Image-text-to-text |
| Context length | Up to 128,000 tokens, subject to runtime settings and available memory |
| Modification | Abliterix; language attention Q/K/V/O projections |
| License | LFM Open License v1.0; see LICENSE |
Abliterix uses layer-specific mean-difference directions from 400 benign and 400 target training prompts, direct orthogonal projection and pre-edit weight normalization.
Weight variants
This repository contains the BF16 model and compressed-tensors quantized variants:
| Location | Precision | Description |
|---|---|---|
| Repository root | BF16 | Complete vision-language model |
FP8/ |
FP8 | Dynamic activation quantization |
Int8/ |
W8A16 | Weight-only, group size 128 |
Int4/ |
W4A16 | Weight-only, group size 128 |
NVFP4/ |
NVFP4A16 | Weight-only, group size 16 |
The vision encoder and multimodal projector remain BF16. Choose a variant for your memory budget and backend; fewer bits do not guarantee lower latency.
Usage
Use the tokenizer, chat template and image processor in the corresponding directory. See the base model card for BF16 Transformers usage and image preprocessing. Quantized checkpoints require compatible LFM3-VL compressed-tensors support; BF16 support alone does not establish support for every quantized format.
Suggested sampling settings: temperature=0.1, top_p=1.0, top_k=50, min_p=0.0, repetition_penalty=1.1.
Evaluation
These metrics describe the BF16 source model; quantized variants have not been separately benchmarked for them.
| Metric | Result |
|---|---|
| Full-vocabulary next-token KL relative to the base model | 0.0426511 |
| StrongREJECT strict refusal rate | 4% |
| StrongREJECT warning while still answering | 38% |
| SimpleSafetyTests strict refusal rate | 8% |
| SimpleSafetyTests warning while still answering | 56% |
KL is mean KL(P_base || P_modified) at one next-token position for each of 100 fixed benign prompts, over all 128,000 vocabulary entries; it is not whole-sequence divergence.
StrongREJECT uses a fixed stratified 100-item subset; SimpleSafetyTests uses all 100 cases. The judge is OllamaCloud deepseek-v4.1-flash:1017. A response that warns but substantively answers is counted as answering, not refusal. Both evaluations had zero generation errors, judge errors, unresolved labels and truncated responses. Free-form generation had no explicit output-token cap, within the context limit.
Abliterix and quantization may affect response quality. Evaluate the exact format, runtime and task before deployment.
License
Derived from LiquidAI/LFM3-VL-2B under its original LFM Open License v1.0.