Replicate

Runway's state of the art video editing model Aleph is on Replicate.
It's like kontext for video.

Transform any video with a prompt: add elements, remove objects, change lighting or weather, generate new camera angles, or change the visual style.

replicate.com/runwayml/ge...

Qwen-Image is on Replicate.
With thanks to Pruna it's just $0.025 per image.
replicate.com/qwen/qwen-i...

We've created some tools to make Veo 3 more useful for you.

- Veo 3 upscaled
- 9:16 output for TikTok/Reels
- JSON prompting

Add these to your apps today: replicate.com/veo-3-apps
(includes Veo 3 Fast)

Wan 2.2 does images too
2 megapixels, ~3 seconds, just $0.02 per image
replicate.com/prunaai/wan...

The era of fast and low cost videos continues.

We're proud to be the first to offer a new faster version of Hailuo 02. Generate 6s and 10s videos in 512p resolution from just $0.10 per video. Incredible quality at this price point.

replicate.com/minimax/hai...

How to prompt Veo 3 with images – Replicate blog You'll be surprised what you can do with AI video now.

This is your ultimate guide to prompting Veo 3 with image.

replicate.com/blog/veo-3-...

Veo 3 now supports image input on Replicate!

Animate any photo.

Try it on Veo 3 or Veo 3 Fast:
replicate.com/google/veo-3
replicate.com/google/veo-...

Our Wan 2.2 models are now even faster and cheaper.

480p is $0.05 per video
720p is $0.10 per video

text-to-video
replicate.com/wan-video/w...

image-to-video:
replicate.com/wan-video/w...

OmniHuman from ByteDance is now on Replicate.
replicate.com/bytedance/o...

Give it an image with a character and some audio, and it'll create a video of them saying those words. It's good with emotion, gestures and can also bring backgrounds to life a little, like the fish in this example.

We have a new FLUX.1 dev model!

Black Forest Labs @bfl_ml and KREA @krea_ai have teamed to create a new image model that solves the "AI look" problem.

Try FLUX.1 Krea [dev] on Replicate today.
replicate.com/black-fores...

We've worked with Pruna AI again to make Wan 2.2 faster, and cheaper.

33 seconds, $0.06 per video, this is the large 14b model at 480p.
replicate.com/wan-video/w... 💨

Wan more?

Replicate and @PrunaAI join forces to bring Wan2.2 to life!

A major upgrade in cinematic quality, smoother movements, and instruction following

replicate.com/wan-video/w...

Veo 3 and Veo 3 Fast both now available in 1080p. No extra cost.

replicate.com/google/veo-3
replicate.com/google/veo-...

replicate/fast-flux-kontext-trainer – Replicate Fine-tune FLUX.1 Kontext to follow your own instructions

Kontext LoRA trainer:
replicate.com/replicate/f...

How to train FLUX.1 Kontext to create a floating car LoRA:

🚨 NEW i2v model on Replicate: Runway Gen-4 Turbo!

Generate studio quality shots with consistent characters, scenes, and cinematic styles. In under 30s.

replicate.com/runwayml/ge...

We've put together a guide on what we think are the best models for creating consistent characters from a single reference image (no lora training required).

replicate.com/blog/genera...

Lipsync-2 by @synclabs_so is now available on Replicate!

Lipsync 2.0 instantly syncs lips to speech with realistic motion and style—no training, just drop in audio.

Test it out: replicate.com/sync/lipsync-2

Here's one we prepared earlier:
replicate.com/fofr/kontex...

The prompt instruction is:
"render this image like a ps1 game (no UI)"

Below is a sample of the training set:

left: start image
right: end image

You can now train FLUX Kontext loras on Replicate.
replicate.com/replicate/f...

- A 1,000 step training takes about 4 minutes
- Train with image pairs, a start and end image
- Write a short prompt instruction that takes you from the start to the end
- Use about 10 image pairs

Try them out here:
replicate.com/bria

Bria is now on Replicate – Replicate blog We've partnered with Bria to bring a suite of commercial-grade image generation and editing models to Replicate. Built entirely on licensed data, Bria’s tools are designed for enterprises and developers building safely with visual AI.

Read about them here:
replicate.com/blog/bria

We've got some more models for you!

Bria AI @bria_ai_ has a suite of commercial-safe image generation + editing tools to try:

- Image 3.2 (image gen)
- Genfill (inpainting)
- Expand image
- Generate background
- Remove background
- Eraser (remove objects)
- Increase resolution

How we optimized FLUX.1 Kontext [dev] – Replicate blog A deep-dive into the Taylor Seer optimization technique

replicate.com/blog/flux-k...

We pride ourselves in remaining open source.
That also means teaching our community.

This is exactly how we optimized FLUX.1 Kontext from Black Forest Labs:

Veo 3 Fast is now on Replicate.

2x faster, 5x cheaper, and same 720p video quality + crisp audio

replicate.com/google/veo-...

It's hard keeping up with all these new video models. We've put together a blog post to help you compare them.

replicate.com/blog/compar...

It covers:

- key specs like price, resolution, duration, fps, speed, and date of release
- features like text-to-video, image-to-video, references, and audio

Hailuo 02 is now live on Replicate
replicate.com/minimax/hai...

Prompt:
"a dog speed climbs up a climbing wall at the olympics"

Runway’s Gen-4 image model is now on Replicate.

replicate.com/runwayml/ge...

Use it to generate images with up to 3 references at once – objects, characters, locations, it all works.

black-forest-labs/FLUX.1-Kontext-dev · Hugging Face

You can get the weights here:
huggingface.co/black-fores...