Install the Transformers runtime stack and run LiquidAI/LFM2.5-VL-3B locally, then send a first image-and-text prompt to confirm the model responds with a generated description.
Install the Transformers runtime stack and run LiquidAI/LFM2.5-VL-3B locally, then send a first image-and-text prompt to confirm the model responds with a generated description.
This setup uses the package documented for this task. Review its source and supported platforms before starting.
The publisher has not provided a complete memory target for this task. Check its hardware guidance before starting.
This setup uses Transformers. The exact checkpoint format, context, and runtime overhead determine its memory need; a weight-file size alone is not a GPU recommendation.
Pick your operating system. Every command below is for the selected package and runtime.
The publisher supplies Python code but does not specify an operating system. Linux is a practical starting environment, not a publisher-tested machine. Run the Python snippets in one session, in order.
On your local machine, install the packages needed to run the model in Transformers: torch, transformers, and torchvision. The model's Quick start is compatible with transformers>=5.0.0, so make sure that version or newer is installed for the Transformers environment you plan to use.
Run this Python setup to load the model on the device in bfloat16 and load its processor. This downloads the model files on first run.
import torch
from transformers import AutoModelForImageTextToText, AutoProcessor
model_id = "LiquidAI/LFM2.5-VL-3B"
model = AutoModelForImageTextToText.from_pretrained(model_id, dtype=torch.bfloat16)
processor = AutoProcessor.from_pretrained(model_id)Prepare an image-plus-text message, tokenize it with the model's chat template, and run generation. The sample image URL keeps the first task self-contained; replace the text with your own prompt when you want a different answer.
messages = [
{
"role": "user",
"content": [
{"type": "image", "url": "https://placecats.com/300/200"},
{"type": "text", "text": "Describe this image."},
],
}
]
inputs = processor.apply_chat_template(
messages,
add_generation_prompt=True,
tokenize=True,
return_dict=True,
return_tensors="pt",
).to(model.device)
with torch.inference_mode():
output_ids = model.generate(
**inputs,
do_sample=True,
temperature=0.2,
top_k=50,
repetition_penalty=1.0,
max_new_tokens=256,
)Decode only the newly generated tokens and print them so you can read the model's first response. A non-empty, on-topic description confirms the model loaded and generated.
generated_ids = output_ids[:, inputs["input_ids"].shape[1] :]
print(processor.batch_decode(generated_ids, skip_special_tokens=True)[0])Ask a short question with a known answer. Confirm the selected local model responds and verify the answer yourself before using it for private work.
This is a source-linked setup, not a YouRunAI hardware test. Confirm your exact runtime version, package, and output before relying on it.
Confirm the exact model files, runtime version, and platform shown in the source-linked steps, then retry the documented command.
Inspect the runtime output, confirm the selected model files loaded, and retry a small known input before changing other settings.
Original instructions, model files, and compatibility notes behind this setup.
Save your machine to see a personalized rating and its reasoning.
Add my hardwareInstall a local runtime, run Qwen3.5 9B, confirm responses, and know when to choose the smaller 4B package.
Connect an open coding-capable model in Ollama to Cline, run a small repository task, and review the result locally.
Use the official FLUX.2 Klein 4B ComfyUI template with exact model files, a first prompt, and an output check.