Instructions to use apple/FastVLM-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use apple/FastVLM-7B with Transformers:

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

pipe = pipeline("text-generation", model="apple/FastVLM-7B", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("apple/FastVLM-7B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use apple/FastVLM-7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "apple/FastVLM-7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "apple/FastVLM-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/apple/FastVLM-7B

SGLang

How to use apple/FastVLM-7B 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 "apple/FastVLM-7B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "apple/FastVLM-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "apple/FastVLM-7B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "apple/FastVLM-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use apple/FastVLM-7B with Docker Model Runner:
```
docker model run hf.co/apple/FastVLM-7B
```

Fix: seamless integration with 🤗 generation pipelines

by sangioai - opened Sep 5, 2025

base: refs/heads/main

←

from: refs/pr/9

Discussion Files changed

-0

Files changed (1) hide show

llava_qwen.py +1 -0

llava_qwen.py CHANGED Viewed

@@ -2186,6 +2186,7 @@ class LlavaQwen2ForCausalLM(Qwen2ForCausalLM, LlavaMetaForCausalLM):
     ) -> Union[GenerateOutput, torch.LongTensor]:
         position_ids = kwargs.pop("position_ids", None)
         attention_mask = kwargs.pop("attention_mask", None)
         if "inputs_embeds" in kwargs:
             raise NotImplementedError("`inputs_embeds` is not supported")

     ) -> Union[GenerateOutput, torch.LongTensor]:
         position_ids = kwargs.pop("position_ids", None)
         attention_mask = kwargs.pop("attention_mask", None)
+        inputs = kwargs.pop("input_ids", inputs)
         if "inputs_embeds" in kwargs:
             raise NotImplementedError("`inputs_embeds` is not supported")