GPT-Image 2.5 Leaks: Luna-Lisa
- OpenAI's new image model "Luna-Lisa" is now being tested on Arena.
- It could be a smaller model, as it appears to be very fast.
- It has a recent knowledge cutoff.
- It shows improvements in realism and overall image quality.
- It could be a smaller version, or a full-fledged model that could be GPT-Image 2.5 or GPT-Image 3, the exact version isn't confirmed yet.
ComfyUI can now split a flat image into separate editable layers and hand you a PSD that opens straight in Photoshop. Thanks to Node 2.0.
The new layer nodes come from Terry (@jtydhr88) on our frontend team. A compositor editor is right in the graph, so you can arrange everything before you ever leave ComfyUI.
How it works:
- Run the Seedream 5.0 layer separation template in Comfy Cloud
- Add multiple images into the layers
- Arrange and composite them in the editor
- Export every layer to a PSD and keep editing in Photoshop
Built on nodes 2.0.
🚨 Grok 4.7 Leak: It's coming
Elon Musk says the next Grok could be something special
> Grok 4.7 reportedly targeting a release in Next weeks
> 2.1T parameters reportedly planned
> Initial training is reportedly already complete
> Massive SpaceX data is being added during supplemental training
> Better token efficiency despite the much larger model
> Slightly slower than Grok 4.6, but reportedly much more capable
> Elon Musk says it could surpass the current frontier
Do you think Grok 4.7 can beat FABLE 5 ?