Jolt-2B

Jolt-2B is a multimodal decision model. Given text or an image and a set of candidate answers, it returns candidate probabilities, a truth estimate, or an ordered score. It does not generate free-form answers.

Model details

Property	Value
Base model	Qwen/Qwen3.5-2B
Base revision	15852e8c16360a2fea060d615a32b45270f8a8fc
Checkpoint	L2, seed 44, selected at update 2,800
Development macro accuracy	82.89% (checkpoint selection partition)
Training objective	Direct-composite loss
Trainable weights	Rank-8 language LoRA and a linear candidate scorer
Input	Text, or text plus one decoded image
Limits	2–128 candidates per question; 4,096-token inference limit

The reported development score selected the checkpoint within its run. It is not an independent test score. Test data were not used to select this checkpoint.

JevBench

18 text dataset benchmarks

8 image dataset benchmarks

You can find a comprehensive model comparison report at: model_comparison_dashboard.html

Use

This repository contains the compact Jolt adapter and inference source. It does not duplicate Qwen's base weights: the runtime downloads the pinned public Qwen revision on first use and caches it locally. The adapter format is Jolt's custom decision-model format, so load it with the included runtime rather than AutoModel.from_pretrained.

The runtime currently requires Python 3.12+, a CUDA-enabled PyTorch setup, and a CUDA GPU. From a clone or downloaded copy of this repository:

python -m pip install -r requirements-cuda.txt
python -m pip install -e . --no-deps

Then:

from jolt.predictor import Predictor

model = Predictor.from_checkpoint("mlengineer-ai/Jolt-2B")
result = model.predict(
    {"message": "I was charged twice. Please refund the duplicate."},
    {
        "route": {
            "type": "choice",
            "instructions": "Which team should handle this request?",
            "criteria": ["billing", "technical support", "sales"],
        },
        "refund": {
            "type": "noul",
            "instructions": "Is a refund requested?",
        },
    },
)

print(result["answers"]["route"]["choice"])
print(result["answers"]["refund"]["noul"])

For an image question, pass a decoded PIL image as state["image"]. This example uses the model loaded above:

from PIL import Image

image = Image.open("path/to/image.jpg").convert("RGB")
image_result = model.predict(
    {"image": image, "question": "Which animal is shown?"},
    {
        "animal": {
            "type": "choice",
            "instructions": "Identify the animal in the image.",
            "criteria": ["cat", "dog", "bird", "other"],
        }
    },
)

print(image_result["answers"]["animal"]["choice"])

Training and data

Jolt follows the recipe based on the Dohnuts project. These checkpoints start from the Qwen3.5 base model listed above; they do not use Dohnuts model weights as their base. Jolt trained rank-8 LoRA weights and a candidate scorer with the base model frozen. Jolt's public-data task suite and split design are adapted from Dohnuts. Its data reference lists source datasets and their known terms. The adapter metadata records fingerprints for Jolt's training, development, calibration, and test partitions. Training data are not included here.

Licensing

This release offers the adapter and decision-head weights under CC BY-NC-SA 4.0; see LICENSE. For this prepared release, this conservative default follows the original Dohnuts model release. It is not a legal determination that every source dataset's terms necessarily apply to trained weights. The training mixture includes sources with different terms, including ScienceQA, whose maintainers specify CC BY-NC-SA 4.0 for the dataset. The available records do not establish commercial clearance for these trained weights; review the source dataset terms before use.

The bundled inference source is adapted from Dohnuts and is licensed under Apache-2.0; see LICENSE-CODE and NOTICE-CODE. The Qwen base model is separately licensed by its publisher under Apache-2.0. Those licenses do not replace or change the terms of training data.