Instructions to use jsun39/Cosine-Beta-KD-Instance with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jsun39/Cosine-Beta-KD-Instance with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="jsun39/Cosine-Beta-KD-Instance")

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("jsun39/Cosine-Beta-KD-Instance", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use jsun39/Cosine-Beta-KD-Instance with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jsun39/Cosine-Beta-KD-Instance"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jsun39/Cosine-Beta-KD-Instance",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/jsun39/Cosine-Beta-KD-Instance

SGLang

How to use jsun39/Cosine-Beta-KD-Instance with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "jsun39/Cosine-Beta-KD-Instance" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jsun39/Cosine-Beta-KD-Instance",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "jsun39/Cosine-Beta-KD-Instance" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jsun39/Cosine-Beta-KD-Instance",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use jsun39/Cosine-Beta-KD-Instance with Docker Model Runner:
```
docker model run hf.co/jsun39/Cosine-Beta-KD-Instance
```

Cosine-Beta-KD-Instance

A 1.7B multimodal LLM checkpoint distilled with Cosine-KD + Beta-KD (Instance-level uncertainty weighting), built on top of MobileVLM with MobileLLaMA-1.4B-Chat as the language backbone.

This checkpoint corresponds to the Beta-KD (Instance) row of the model zoo in Beta-KD: Uncertainty-Aware Knowledge Distillation for Multimodal Large Language Models.

Model Details

Item	Value
Architecture	MobileVLM (CLIP visual encoder + LDP projector + MobileLLaMA LLM)
Language model	MobileLLaMA 1.4B
Distillation losses	Cosine-KD (logit alignment) + Beta-KD instance-level uncertainty loss
Training step	`checkpoint-18000`
Total params	~1.7B
Precision	fp16

Evaluation

Evaluated on six standard multimodal benchmarks (no beam search, greedy decoding to match the chat-demo behavior).

Method	LLM	MME^P	MME^A	GQA	VQA^T	POPE	MMB^dev	SQA^I	Avg.
Cosine-KD baseline	MobileLLaMA 1.4B	1308.4	65.4	59.9	52.2	84.6	57.1	61.3	63.4
+ Beta-KD (Task)	MobileLLaMA 1.4B	1352.0	67.6	60.8	53.9	85.4	59.1	61.2	64.7
+ Beta-KD (Instance) (this model)	MobileLLaMA 1.4B	1350.3	67.5	61.2	54.2	86.0	60.2	62.9	65.3

Usage

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

repo_id = "jsun39/Cosine-Beta-KD-Instance"

tokenizer = AutoTokenizer.from_pretrained(repo_id, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    repo_id,
    torch_dtype=torch.float16,
    trust_remote_code=True,
).cuda()

For full inference (image + text), please follow the inference example in the Beta-KD repo — the visual encoder / projector loading, image preprocessing, and chat template are described there.

Files

This repo contains only the files needed for inference:

pytorch_model.bin — fp16 weights
config.json, generation_config.json
tokenizer.model, tokenizer_config.json, special_tokens_map.json

DeepSpeed optimizer / RNG / trainer states are intentionally not uploaded.

Citation

@article{sun2026betakd,
  title   = {Beta-KD: Uncertainty-Aware Knowledge Distillation for Multimodal
             Large Language Models},
  author  = {Sun, Jingchen and Han, Shaobo and Patel, Deep and Kohno, Wataru and Jin, Can and Chen, Changyou},
  journal = {CVPR},
  year    = {2026}
}

License

Released under the Apache-2.0 license, inheriting from MobileVLM and MobileLLaMA. The visual encoder and any third-party data follow their original licenses.