Palet logo

People-powered cloud platform for generative AI.

palet icon

Built on an open network of
inference servers

servers

Community-driven models

Utilize open source models through a network of community-owned servers for widespread access.

Scalable and affordable

Deploy inference servers globally without data centers using a BitTorrent-like network for scalable model distribution.

Multi modal execution

Execute various models simultaneously, covering vision, language, and speech, across a network.

Run over 100+ open models or
Deploy your own models

https://palet.app
Upload model icon

LLAMA-2

Chat

Mixtral Instruct

Chat

Striped Hyena Nous

Chat

Mistral instruct

Chat

01-AI YI

Language

UAE-LARGE V1

Embeddings

Stable Diffusion XL 1.0

Image

M2-BERT 80M 32K Retrieval

Embeddings

Code LLAMA Instruct

Chat

REDPAJAMA-INCITE INSTRUCT

Language

Vicuna V1.5 16K

Chat

NOUS CAPYBARA

Chat

Nexus Raven

Language

NOUS HERMES LLAMA-2

Chat

WIZAR LM

Chat

Wizardcoder Python v1.0

Code

Features at glance

Stream inference from any open source model or augment your on-prem server.

Circumvent censorship

Access hundreds of open models or deploy yours on Palet's network, designed to resist censorship and ensure your server remains online.

Bring your own server

Integrate your server to cut costs and manage core tasks locally, while expanding reach and distribution with our global network.

7x Lower cost

Achieve substantial savings with LLaMA-2 70B models compared to GPT-3.5 TURBO, offering cost-effective scalability.

Start Building

Get started in minutes by creating an account and get $50 in free API credits.