We've pushed a new version that fixes this bug https://
replicate.com/cjwbw/tokenflo
w/versions/d456a26c0c1764ac4d9b933a564fe70e660c46423ba0afb5d3200f1a46060728
…
TOOLS
-
TokenFlow Bug Fix Release New Version
By
–
-
TokenFlow: AI-Powered Video Editing with Text Prompts
By
–
Use TokenFlow to edit a video with a prompt:https://t.co/eSJ9qkFoAi
— fofr (@fofrAI) 23 janvier 2024
Here's an example where I used prompts to turn:
– the video into an animation
– the wolf into a chrome robot
It uses DDIM inversion and a Stable Diffusion base weight. Paper: https://t.co/rgBSOlMHE1 pic.twitter.com/FCm8jgMq9bUse TokenFlow to edit a video with a prompt: https://
replicate.com/cjwbw/tokenflow Here's an example where I used prompts to turn: – the video into an animation
– the wolf into a chrome robot It uses DDIM inversion and a Stable Diffusion base weight. Paper: https://
diffusion-tokenflow.github.io -
Signal: Privacy-First Communication Without Data Collection
By
–
Signal is by far the most private/secure (& open/audited for those wanting proof). We collect no data about you & can't turn any over. Signal calls don't require text. A threat model where text (interlocutors) isn't trusted is valid, but this isn't a prob w Signal or its security
-
Perplexity Outperforms Grok for AI Research Applications
By
–
I suggest using Perplexity by @AravSrinivas for research. It’s way better for that use case. Grok is cool and it’s amazing how fast Elon has made progress. But they’re still very far behind the best models from OpenAI. Which Perplexity is built on top of.
-
Perplexity’s AI-First Browser Strategy Keeps Users Inside App
By
–
One of the smartest things @perplexity_ai is doing is keeping browsing inside the app. You search something.
Get sources.
Click. It doesn’t open Safari, it’s in Perplexity. If you screenshot, it knows. If you click more sites, it knows. More like an AI-first browser, not -
InstantID Now Available on Replicate Platform
By
–
InstantID up on Replicate 🙌
— fofr (@fofrAI) 22 janvier 2024
Fast work @zsakib_ https://t.co/FZQTxZIbFlInstantID up on Replicate Fast work @zsakib_
-
TEL launches Model Lab tutorial for training inference cost optimization
By
–
Our Japanese partner TEL has put together a great English tutorial of Model Lab, a tool we built to help you balance training vs. inference costs for different model/data sizes. Check it out!
-
Playground v2 Image Generation Model Now Available for Bot Building
By
–
You can now build image generating bots using Playground v2 as your base model!
— Poe (@poe_platform) 22 janvier 2024
Playground v2 is a powerful image model that delivers cutting-edge visuals with a focus on aesthetic quality and impressive accuracy. We’re excited to see the bots you build on it! pic.twitter.com/8utVxOzn56You can now build image generating bots using Playground v2 as your base model! Playground v2 is a powerful image model that delivers cutting-edge visuals with a focus on aesthetic quality and impressive accuracy. We’re excited to see the bots you build on it!
-
Databricks Admin Experience Enhanced with Cloud Integrations
By
–
The Databricks admin experience just got a glow-up What to expect: Streamlined integrations with Power BI & @tableau for Databricks Unified single sign-on for Databricks & @awscloud accounts OAuth support through @Azure and @Google tokens