Blog

Wan2.1: generate videos with an API

Wan2.1 is the most capable open-source video generation model, producing coherent and high-quality outputs. Learn how to run it in the cloud with a single line of code.

Wan2.1 parameter sweep

We've been playing with Alibaba's WAN2.1 text-to-video model lately. What happens when you tweak those mysterious parameters? Let's find out.

You can now fine-tune open-source video models

Train your own versions of Tencent's HunyuanVideo for style, motion, and characters on Replicate.

Generate short videos with the Replicate playground

Create AI videos with a convenient workflow.

AI video is having its Stable Diffusion moment

There are lots of models that are as good as OpenAI's Sora now.

FLUX fine-tunes are now fast

We've made running fine-tunes on Replicate much faster, and the optimizations are open-source.

FLUX.1 Tools – Control and steerability for FLUX

A new set of image generation capabilities for FLUX models, including inpainting, outpainting, canny edge detection, and depth maps.

NVIDIA L40S GPUs are here

NVIDIA L40S GPUs are here, with better performance and lower cost.

We messed up: data URLs in our sync API

We've decided to stop returning data URLs in sync API outputs based on feedback from users. We're going to take a few steps back and figure out the best way to get your model output as fast as possible.

Ideogram v2 is an outstanding new inpainting model

We've partnered with Ideogram to bring their inpainting model to Replicate's API.

Stable Diffusion 3.5 is here

Stability AI's latest text-to-image model is now available on Replicate and you can run it with an API.

FLUX is fast and it's open source

FLUX is now much faster on Replicate, and we’ve made our optimizations open-source so you can see exactly how they work and build upon them.

FLUX1.1 [pro] is here

Black Forest Labs continue to push boundaries with their latest release of FLUX.1 image generation model.

Using synthetic training data to improve Flux finetunes

It's easy to fine-tune Flux, but sometimes you need to do a little more work to get the best results. This post covers techniques you can use to improve your fine-tuned Flux models.

Fine-tune FLUX.1 with an API

Create and run your own fine-tuned Flux models programmatically using Replicate's HTTP API.

Fine-tune FLUX.1 to create images of yourself

Create your own fine-tuned Flux model to generate new images of yourself.

Replicate Intelligence #12

Flux LoRAs, Hot Zuck, and Replicate on Lex Fridman

Replicate Intelligence #11

Fine tune FLUX.1, generative video games, a vision for the metaverse

Fine-tune FLUX.1 with your own images

We've added fine-tuning (LoRA) support to FLUX.1 image generation models. You can train FLUX.1 on your own images with one line of code using Replicate's API.

Replicate Intelligence #10

Flux developments, Minecraft bot, Streamlit cookbook with Zeke

FLUX.1: First Impressions

We explore FLUX.1's unique strengths and aesthetics to see what we can generate.

Replicate Intelligence #9

Open source frontier image model, cut objects from videos, new Python web framework from Jeremy Howard

Run FLUX with an API

FLUX.1 is a new text-to-image model from Black Forest Labs, the creators of Stable Diffusion, that exceeds the capabilities of previous open-source models.

Replicate Intelligence #8

A top-tier open-ish language model, new safety classifiers, model search API

Run Meta Llama 3.1 405B with an API

Llama 3.1 405B: is the most powerful open-source language model from Meta. Learn how to run it in the cloud with one line of code.

Replicate Intelligence #7

Data curation, data generation, data data data

Replicate Intelligence #6

Google's Gemma2 models, language model leaderboard, tips for Stable Diffusion 3

Replicate Intelligence #5

Really good coding model, AI search breakthroughs, Discord support bot

How to get the best results from Stable Diffusion 3

We show you how to use Stable Diffusion 3 to get the best images, including new techniques for prompting.

Run Stable Diffusion 3 on your Apple Silicon Mac

A step-by-step guide to generating images with Stable Diffusion 3 on your M-series Mac using MPS acceleration.

Push a custom version of Stable Diffusion 3

Create your own custom version of Stability's latest image generation model and run it on Replicate via the web or API.

Replicate Intelligence #4

Find concepts in GPT models, real-time speech to text in the browser, H100s are coming

Run Stable Diffusion 3 on your own machine with ComfyUI

Copy and paste a few commands into terminal to play with Stable Diffusion 3 on your own GPU-powered machine.

H100s are coming to Replicate

We'll soon support NVIDIA's H100 GPUs for predictions and training. Let us know if you want early access.

Run Stable Diffusion 3 with an API

Stable Diffusion 3 is the latest text-to-image model from Stability, with improved image quality, typography, prompt understanding, and resource efficiency. Learn how to run it in the cloud with one line of code.

Replicate Intelligence #3

Garden State Llama, applied LLMs guide, real-time image generation

Replicate Intelligence #2

Faster image generation, AI-powered world simulator, insights on AI dataset complexity

Replicate Intelligence #1

DIY Llama 3 implementation, open-source smart glasses, steering language models with dictionary learning

Shared network vulnerability disclosure

Run Snowflake Arctic with an API

Arctic is a new open-source language model from Snowflake. Learn how to run it in the cloud with one line of code.

Run Meta Llama 3 with an API

Llama 3 is the latest language model from Meta. Learn how to run it in the cloud with one line of code.

Run Code Llama 70B with an API

Code Llama 70B is one of the powerful open-source code generation models. Learn how to run it in the cloud with one line of code.

How to create an AI narrator for your life

Or, how I met a virtual David Attenborough.

Clone your voice using open-source models

We’ve added fine-tuning for realistic voice cloning (RVC). You can train RVC on your own dataset from a YouTube video with a few lines of code using Replicate's API.

Businesses are building on open-source AI

We've raised a $40 million Series B led by a16z.

How to run Yi chat models with an API

The Yi series models are large language models trained from scratch by developers at 01.AI. Learn how to run them in the cloud with one line of code.

Scaffold Replicate apps with one command

We've added a CLI command that makes it easy to get started with Replicate.

Using open-source models for faster and cheaper text embeddings

An interactive example showing how to embed text using a state-of-the-art embedding model that beats OpenAI's embeddings API on price and performance.

Generate music from chord progressions and text prompts with MusicGen-Chord

We’ve added chord conditioning to Meta’s MusicGen model, so you can create automatic backing tracks in any style using text prompts and chord progressions.

Generate images in one second on your Mac using a latent consistency model

How to run a latent consistency model on your M1 or M2 Mac

How to use retrieval augmented generation with ChromaDB and Mistral

In this post we'll explore the basics of retrieval augmented generation by creating an example app that uses bge-large-en for embeddings, ChromaDB for vector store, and mistral-7b-instruct for language model generation.

Fine-tune MusicGen to generate music in any style

We’ve added fine-tuning support to MusicGen. You can train the small, medium and melody models on your own audio files using Replicate.

Jet-setting with Llama 2 + Grammars

How to use Llama 2 models with grammars for information extraction tasks.

How to run Mistral 7B with an API

Mistral 7B is an open-source large language model. Learn what it's good at and how to run it in the cloud with one line of code.

Make smooth AI generated videos with AnimateDiff and an interpolator

Combine AnimateDiff and the ST-MFNet frame interpolator to create smooth and realistic videos from a text prompt

Fine-tuned models now boot in less than one second

We've made some dramatic improvements to cold boots for fine-tuned models.

Painting with words: a history of text-to-image AI

With the recent release of Stable Diffusion XL fine-tuning on Replicate, and today being the 1-year anniversary of Stable Diffusion, now feels like the perfect opportunity to take a step back and reflect on how text-to-image AI has improved over the last few years.

We're cutting our prices in half

The price of public models is being cut in half, and soon we'll start charging new users for setup and idle time on private models.

A guide to prompting Llama 2

Learn the art of the Llama prompt.

Streaming output for language models

Our API now supports server-sent event streams for language models. Learn how to use them to make your apps more responsive.

Fine-tune SDXL with your own images

We’ve added fine-tuning (Dreambooth, Textual Inversion and LoRA) support to SDXL 1.0. You can train SDXL on your own images with one line of code using the Replicate API.

Run Llama 2 with an API

Llama 2 is the first open source language model of the same caliber as OpenAI’s models. Learn how to run it in the cloud with one line of code.

Run SDXL with an API

How to run Stable Diffusion XL 1.0 using the Replicate API

A comprehensive guide to running Llama 2 locally

How to run Llama 2 on Mac, Linux, Windows, and your phone.

Fine-tune Llama 2 on Replicate

So you want to train a llama...

What happened with Llama 2 in the last 24 hours? 🦙

A roundup of recent developments from the llamaverse following the second major release of Meta's open-source large language model.

Make any large language model a better poet

Prompt engineering and training are often the first solutions we reach for to improve language model behavior, but they're not the only way.

Status page

We've added a status page to provide real-time updates on the health of Replicate.

Language model roundup, April 2023

A roundup of recent developments from the world of open-source language models.

AutoCog — Generate Cog configuration with GPT-4

Give it a machine learning directory and AutoCog will create predict.py and cog.yaml until it successfully runs a prediction

Language models are on Replicate

How to use Alpaca-LoRA to fine-tune a model like ChatGPT

Week 3 of LLaMA 🦙

A roundup of recent developments from the llamaverse.

Fine-tune LLaMA to speak like Homer Simpson

With a small amount of data and an hour of training you can make LLaMA output text in the voice of the dataset.

Train and run Stanford Alpaca on your own machine

We'll show you how to train Alpaca, a fine-tuned version of LLaMA that can respond to instructions like ChatGPT.

Machine learning needs better tools

Lots of people want to build things with machine learning, but they don't have the expertise to use it.

Introducing LoRA: A faster way to fine-tune Stable Diffusion

It's like DreamBooth, but much faster. And you can run it in the cloud on Replicate.

Train and deploy a DreamBooth model on Replicate

With just a handful of images and a single API call, you can train a model, publish it to Replicate, and run predictions on it in the cloud.

Run Stable Diffusion on your M1 Mac’s GPU

How to run Stable Diffusion locally so you can hack on it

Run Stable Diffusion with an API

How to use Replicate to integrate Stable Diffusion into hacks, apps, and projects

Build a robot artist for your Discord server with Stable Diffusion, Replicate, and Fly.io

A tutorial for building a chat bot that replies to prompts with the output of a text-to-image model.

Join us at Uncanny Spaces

We're bringing people together to explore what's being created with machine learning.

Automating image collection

Using CLIP and LAION5B to collect thousands of captioned images.

Exploring text to image models

The basics of using the API to create your own images from text.

A new template for model READMEs

Inspired by model cards, we've created templates for documenting models on Replicate.

Constraining CLIPDraw

An introduction to differentiable programming and the process of refining generative art models.

Hello, world!

We're a small team of engineers and machine learning enthusiasts working to make machine learning more accessible.