You can now fine-tune MusicGen to generate music in any style. – Train a fine-tune in 15 mins
– Automatically labels your training data
– Automatically removes vocals
– Makes a track as long as you want
@replicate
-
MusicGen Fine-Tuning: Create Custom Style Music in Minutes
By
–
-
Llava Image Chat Support Added to Llama2.ai
By
–
https://t.co/3ZZemXeQjs pic.twitter.com/bAYkUniXwu
— Replicate (@replicate) 12 octobre 2023I just added Llava support to http://
llama2.ai — you can now chat with your images! It's fully open-source and pretty amazing. -
AnimateDiff and Interpolator Create Smooth Text-to-Video
By
–
We wrote about how to use AnimateDiff and an interpolator to create smooth videos from text prompts.
— Replicate (@replicate) 7 octobre 2023
📹 Reply with the video you make!https://t.co/9cDSnuONTO pic.twitter.com/rWSihXoWwEWe wrote about how to use AnimateDiff and an interpolator to create smooth videos from text prompts. Reply with the video you make! https://
replicate.com/blog/animatedi
ff-interpolator
… -
New API for Discovering Public Models on Replicate
By
–
We've added a new API for listing public models. You can use it to discover newly published models and to build your own tools for exploring the thousands of open-source models published by our community. https://
replicate.com/changelog/2023
-10-05-api-for-listing-public-models
… -
Replicate launches AI hype tracking tool
By
–
Having trouble keeping up with the AI hype?
— Replicate (@replicate) 4 octobre 2023
Introducing https://t.co/567AHyg1ey pic.twitter.com/c1iakBxL7uHaving trouble keeping up with the AI hype? Introducing http://
hype.replicate.dev -
Real-time model throughput and autoscaling monitoring capabilities
By
–
You can monitor model throughput and autoscaling in real time.
-
Roll Out New Model Versions Without Downtime or Code Changes
By
–
You can roll out new versions of the model you're running, so you can change versions without downtime and without having to edit your code.
-
Replicate Deployments: Control, Visibility, and Model Management
By
–
It's easy to run models on Replicate, but what if you need to keep a model switched on, roll out a new version, or monitor how it's performing? You can now create a deployment to give you more control and visibility.
-
Mistral 7B Now Available on Replicate, Outperforms Llama 2
By
–
Mistral 7B is live on Replicate. It outperforms Llama 2 13B on all benchmarks we tested. Instruction-tuned model: https://
replicate.com/a16z-infra/mis
tral-7b-instruct-v0.1
… Base model: https://
replicate.com/a16z-infra/mis
tral-7b-v0.1
…
