How do I talk to GPT?

Note: This page is about typing text to GPT. If you are looking for how you can use your voice to speak to GPT-3 and hear its responses, please see my article on using the macOS CLI + dictation + TTS (say).

Note: Most large language models are not chatbots, but they can be set up to emulate conversations as one of their many, many capabilities.

Update 30/Nov/2022: It is now possible to talk with OpenAI’s ChatGPT, a version of GPT-3.5 fine-tuned on dialogue. Chat for free using OpenAI’s playground: chat.openai.com. The model is hamstrung and limited for safety, using several rules and policies much like DeepMind Sparrow. Watch the demonstration of ChatGPT livestreamed by Alan on 30/Nov/2022.

Want to talk to BlenderBot 3.0?

Launched 6/Aug/2022. US only.

BlenderBot 3.0 chatbot demo.
This chatbot model was launched by Meta AI in Aug/2022, and is the same size as GPT-3, but is a specialized chatbot with full conversational memory and internet access, based on Meta AI’s OPT-175B.
Web: blenderbot.ai (US only)


Want to talk to GPT-3?

Launched May 2020. Requires SMS authentication.

Emerson. Via Telegram or iOS or Android…
Emerson (by Quickchat.ai) is a fully conversational and Multilingual AI Assistant powered by OpenAI’s GPT-3, a 175 billion parameter state-of-the-art language model.
Telegram: https://www.quickchat.ai/emerson
iOS: App store download
Android: Play store download

For a discount code on the GPT-3 powered Emerson AI by Quickchat.ai (in Telegram), join The Memo and receive a code in your first email!

Playground. Via web, using the original GPT-3 model playground hosted by OpenAI…
https://beta.openai.com/playground
Set your playground up via prompt crafting for talking with Leta AI

Copyhat. Via web or iPhone or Android…
Copyhat uses GPT-3 175B (Davinci) in their ‘Ask Philosopher’ module.
https://copyhat.com/


Want to talk to GPT-NeoX-20B?

Launched Feb 2022.

Playground. Via web, using a GPT-NeoX-20B model playground hosted by Riku.ai…
https://riku.ai/dashboard/playground

Playground. Via web, using a GPT-NeoX-20B model playground hosted by Forefront…
https://playground.helloforefront.com/models/free-gpt-neox-playground

EleutherAI. Via web, using the original GPT-NeoX-20B model demo playground hosted by EleutherAI…
https://20b.eleuther.ai/


Want to talk to GPT-J?

Launched June 2021.

Forefront. Via web, using a faster playground for the GPT-J-6B model hosted by Forefront…
https://playground.helloforefront.com/

EleutherAI. Via web, using the original GPT-J-6B model demo playground hosted by EleutherAI…
https://6b.eleuther.ai/


Want to talk to Jurassic-1, the largest and most sophisticated language model ever released for general use by developers?

Launched August 2021. Requires SMS authentication.

Studio. Via web, using the original Jurassic-1 model studio hosted by AI21 in Israel…
https://studio.ai21.com/


Want to talk to something simple, without downloading anything or signing up?

Launched April 2020.

Inferkit. Via web, write with Megatron-11B (based on RoBERTa) by FB AI, hosted by InferKit. Press TAB to have the model continue your sentence…
https://app.inferkit.com/demo


Want to build it yourself?

Web: Check out James Weaver’s Talk w/GPT-3 repo on GitHub.


Want even more options?


Summary of current models: View the full data (Google sheets)
Download PDF version
Visit LifeArchitect.ai/models for a full list of public playgrounds and platforms.


Due to licensing (and its size), you cannot download GPT-3, only access through an API with an approvals process. There are options for downloading and running GPT-J with low VRAM (suggest 16GB RAM + 8GB GPU).

Get The Memo

by Dr Alan D. Thompson · Be inside the lightning-fast AI revolution.
Hundreds of paid subscribers. Readers from Microsoft, Tesla, & Google AI.
Artificial intelligence that matters, as it happens, in plain English.
Get The Memo.


Dr Alan D. Thompson is an AI expert and consultant, advising Fortune 500s and governments on post-2020 large language models. His work on artificial intelligence has been featured at NYU, with Microsoft AI and Google AI teams, at the University of Oxford’s 2021 debate on AI Ethics, and in the Leta AI (GPT-3) experiments viewed more than 2.5 million times. A contributor to the fields of human intelligence and peak performance, he has held positions as chairman for Mensa International, consultant to GE and Warner Bros, and memberships with the IEEE and IET. He is open to consulting and advisory on major AI projects with intergovernmental organizations and enterprise.

This page last updated: 2/Dec/2022. https://lifearchitect.ai/how-do-i-talk-to-gpt/