.tryit-button{
  display: none !important;
}

.centered-table {
  text-align: center;
}

.centered-table table {
  margin: 0 auto !important;
  display: table !important;
}

/* Alternative approach - target all tables in centered-table divs */
div.centered-table > table {
  margin-left: auto !important;
  margin-right: auto !important;
}
@media (prefers-color-scheme: dark) {
  .gradient-background {
    background: linear-gradient(to bottom, #000000, #000000) !important;
  }
}

.mint-absolute {
    position: absolute
}
.mint-relative {
    position: relative
}
.mint-inset-0 {
    inset: 0px
}
.mint-ml-1 {
    margin-left: 0.25rem
}
.mint-ml-1\.5 {
    margin-left: 0.375rem
}
.mint-mt-3 {
    margin-top: 0.75rem
}
.mint-mt-4 {
    margin-top: 1rem
}
.mint-block {
    display: block
}
.mint-flex {
    display: flex
}
.mint-hidden {
    display: none
}
.mint-h-4 {
    height: 1rem
}
.mint-h-\[18px\] {
    height: 18px
}
.mint-w-4 {
    width: 1rem
}
.mint-w-\[18px\] {
    width: 18px
}
.mint-w-full {
    width: 100%
}
.mint-flex-none {
    flex: none
}
.mint-flex-col {
    flex-direction: column
}
.mint-flex-wrap {
    flex-wrap: wrap
}
.mint-items-center {
    align-items: center
}
.mint-justify-center {
    justify-content: center
}
.mint-gap-3 {
    gap: 0.75rem
}
.mint-rounded-full {
    border-radius: 9999px
}
.mint-border {
    border-width: 1px
}
.mint-bg-white {
    --tw-bg-opacity: 1;
    background-color: rgb(255 255 255 / var(--tw-bg-opacity))
}
.mint-py-4 {
    padding-top: 1rem;
    padding-bottom: 1rem
}
.mint-pl-4 {
    padding-left: 1rem
}
.mint-pr-4 {
    padding-right: 1rem
}
.mint-text-sm {
    font-size: 0.875rem;
    line-height: 1.25rem
}
.mint-leading-6 {
    line-height: 1.5rem
}
.mint-text-\[\#212121\] {
    --tw-text-opacity: 1;
    color: rgb(33 33 33 / var(--tw-text-opacity))
}
.mint-text-\[\#52525c\] {
    --tw-text-opacity: 1;
    color: rgb(82 82 92 / var(--tw-text-opacity))
}
.mint-text-gray-500 {
    --tw-text-opacity: 1;
    color: rgb(107 114 128 / var(--tw-text-opacity))
}
.mint-text-gray-900 {
    --tw-text-opacity: 1;
    color: rgb(17 24 39 / var(--tw-text-opacity))
}
.mint-transition-colors {
    transition-property: color, background-color, border-color, text-decoration-color, fill, stroke;
    transition-timing-function: cubic-bezier(0.4, 0, 0.2, 1);
    transition-duration: 150ms
}
.hover\:mint-bg-gray-600:hover {
    --tw-bg-opacity: 1;
    background-color: rgb(75 85 99 / var(--tw-bg-opacity))
}
.dark\:mint-text-gray-200:is(.dark *) {
    --tw-text-opacity: 1;
    color: rgb(229 231 235 / var(--tw-text-opacity))
}
.dark\:mint-text-white:is(.dark *) {
    --tw-text-opacity: 1;
    color: rgb(255 255 255 / var(--tw-text-opacity))
}
.dark\:hover\:mint-bg-white\/70:hover:is(.dark *) {
    background-color: rgb(255 255 255 / 0.7)
}
@media (min-width: 1024px) {
    .lg\:mint-flex {
        display: flex
    }
}

AI21 Studio

Overview

AI21 offers Python and TypeScript libraries that simplify the process of using our LLMs' API.

Introducing the Jamba Family of Open Models.

Jamba

Model Availability by Platform

Prompt Engineering for Jamba Models

Build conversational experiences that interact with your organizational documents.

Conversational RAG

AI21 Maestro is in beta, available exclusively to early access users.

Deploy AI21's Jamba models using vLLM in your own environment. vLLM is an open-source library for high-throughput LLM inference and serving.

vLLM

Deploy AI21's Jamba models on managed cloud services for production workloads

Cloud Platform Deployment

Resolve common issues and optimize performance for AI21's Jamba model deployments

Troubleshooting & Performance Optimization

Fine-tuning is the process of adapting a pre-trained model to perform better on specific tasks by training it on domain-specific data. Learn how to fine-tune Jamba models using different approaches including full fine-tuning, LoRA, and QLoRA

Fine-tuning

Quantization reduces model memory usage by representing weights with lower precision. Learn how to use quantization techniques with Jamba models for efficient inference and training.

Quantization

Tokenization is the process of converting text into numerical tokens that language models can understand. Learn how to use AI21's tokenizer for Jamba models with practical examples.

Tokenization

Account

Pricing

Responsible Use

Safety Research

Training Data

Status

Community

Blog

Introduction

Authentication

Rate Limits

Chat Request

Chat Response

File Library

Upload Workspace Files

Retrieve a list of documents in the user's library.  You can optionally specify a filter to return only files with matching labels or paths.

List Library Files

Get metadata about a specific file by file ID. The file ID is generated by AI21 when you upload the file.

Get Workspace Files

Get a link used to download the file contents from the library.

Get File Download Link

Update the specified parameters of a specific document in the user's library. This operation currently supports updating the publicUrl and labels parameters.

Update File

Delete the specified file from the library.

Delete File

Build conversational experiences that interact with your organizational documents

AI21 Maestro is currently in beta and available exclusively to early access users.

Create run

Retrieves the status of the run execution based on its ID. The status can be `completed`, `failed`, or `in_progress`.

Terms and Policies

Status Page

Trust Center

Terms of Service

Privacy Policy

Publications

Blog

Press