How to deploy google/flan-ul2 - simple. (open source ChatGPT alternative)

Published on 2023.03.17 by Nikola Borisov

Flan-UL2 is probably the best open source model available right now for chatbots. In this post we will show you how to get started with it very easily. Flan-UL2 is large - 20B parameters. It is fine tuned version of the UL2 model using Flan dataset. Because this is quite a large model it is not easy to deploy it on your own machine. If you rent a GPU in AWS, it will cost you around $1.5 per hour or $1080 per month. Using DeepInfra model deployments you only pay for the inference time, and we do not charge for cold starts. Our pricing is $0.0005 per second of running inference on Nvidia A100. Which translates to about $0.0001 per token generated by Flan-UL2.

Also check out the model page https://deepinfra.com/google/flan-ul2. You can run inferences, check the docs/API for running inferences via curl.

Getting started

First, you'll need to get an API key from the DeepInfra dashboard.

Sign up or log in to your DeepInfra account
Navigate to the API Keys section in the dashboard
Create a new API key for authentication

Deployment

You can deploy the google/flan-ul2 model easily through the web dashboard or API. The model will be automatically deployed when you first make an inference request.

Inference

You can use it with our REST API. Here's how to use it with curl:

curl -X POST \
    -d '{"prompt": "Hello, how are you?"}' \
    -H 'Content-Type: application/json' \
    -H "Authorization: Bearer YOUR_API_KEY" \
    'https://api.deepinfra.com/v1/inference/google/flan-ul2'
copy

To see the full documentation of how to call this model, check out the model page on the DeepInfra website or the API documentation.

If you want a list of all the models you can use on DeepInfra, you can visit the models page on our website or use the API to get a list of available models.

There is no easier way to get started with arguably one of the best open source LLM. This was quite easy right? You did not have to deal with docker, transformers, pytorch, etc. If you have any question, just reach out to us on our Discord server.

Related articles

DeepInfra Raises $107M Series B to Scale Inference InfrastructureDeepInfra has raised $107 million in Series B funding to scale its inference cloud, expand global capacity, and support the next generation of open-source and agentic AI workloads.

GLM-5.3 API Providers: Speed, Latency & Cost<p>GLM-5.3 API Review Summary GLM-5.3 is Z AI’s flagship coding and agentic reasoning model, released on August 14, 2026, and available on DeepInfra since launch. The model is built on the same ~753-billion parameter Mixture of Experts (MoE) base architecture as GLM-5.2, with all performance gains derived entirely from scaled post-training rather than architectural changes. […]</p>

DeepSeek V4 Flash vs Qwen3.6 vs GLM-4.6 Benchmarks<p>A breakdown of three open-weight models across intelligence, speed, and inference cost. Three open-weight models cover most of what a developer needs from open inference right now: DeepSeek V4 Flash, Qwen3.6 35B A3B, and GLM-4.6. All three run on DeepInfra, and all three use a Mixture-of-Experts design that keeps active parameters low while total capacity […]</p>

View all