// free test deployments

Paste a Hugging Face link.
Get a working endpoint.

Pilot, our AI agent, reads the model, picks a GPU, deploys it and gives you an OpenAI-compatible API. Free test launches last 15 minutes and 25 requests.

continue with litepod account →

pilota few minutes

$ modeldock launch Qwen/Qwen3-4B
✓ researching model       text-generation, 4B
✓ selecting GPU           24 GB pool
✓ deploying               vLLM worker
✓ warm                    ready
→ POST https://api.modeldock.dev/v1/chat/completions█

Paste a link

Any open Hugging Face model up to 10B parameters. Gated models are not supported.

Pilot launches it

Pilot, our AI agent, checks the model, sizes a GPU and deploys it. Typically a few minutes.

Call it like OpenAI

Your model gets its own id on an OpenAI-compatible endpoint. Use your existing SDK code.

// deploy your own model

Deploy any Hugging Face model, free to try

Paste a Hugging Face link, up to 10B parameters. Pilot, our AI agent, launches it as a free test deployment with its own model id, behind an OpenAI-compatible endpoint.

client.chat.completions.create(
  model="modeldock/qwen3-8b-…",
  messages=[…]
)

Alibaba Qwenready to launch

Qwen3 0.6B

modeldock/qwen3-0.6b-…

0.8B params · free to test

OpenBMBready to launch

MiniCPM5 2B

modeldock/minicpm5-2b-…

2.5B params · free to test

LiquidAIready to launch

LFM2.5 2.6B

modeldock/lfm2.5-2.6b-…

2.7B params · free to test

up to 10B parameterssame OpenAI-compatible APIfree test deployment

Deploy your first model free

open modeldock →

// trending on hugging face

Launch a model people love

Popular open models from Hugging Face, up to 10B parameters. Paste any link and Pilot does the rest.

MODEL	LAB	PARAMS	DOWNLOADS	NOTES
Qwen3 0.6B	Alibaba Qwen	0.8B	30.8M	Trending now, 1.7K likes on Hugging Face.
MiniCPM5 2B	OpenBMB	2.5B	1.2M	Trending now, 1.7K likes on Hugging Face.
LFM2.5 2.6B	LiquidAI	2.7B	104K	Trending now, 799 likes on Hugging Face.
Qwen3 8B	Alibaba Qwen	8.2B	11.7M	Strong all-rounder with a thinking mode.
Qwen3 4B	Alibaba Qwen	4.0B	5.8M	Small, quick and surprisingly capable.
Qwen2.5 7B Instruct	Alibaba Qwen	7.6B	9.6M	A go-to chat and instruction model.
DeepSeek R1 Distill 7B	DeepSeek	7.6B	270K	Step-by-step reasoning in a small model.
Mistral 7B Instruct	Mistral AI	7.2B	2.2M	The classic open 7B, with tool calling.

// updated daily from huggingface.co

Paste a Hugging Face link.Get a working endpoint.