Black Forrest is increasingly following the path of Pika Labs or Higgsfield.
They launched "Kontext Komposer" and "Kontext-powered Presets
Now you can upload your image and ask to change it by simply selecting a preset, rather than writing out a complex prompt. A one-button solution for those who don't know how\don't want to make prompts.
It's clear that under the hood there is some kind of system prompt for each preset (which LLM I don't know). So smart people first dug up all these prompts, then screwed them into Komfi and even tried them in other generators\chats.
They launched "Kontext Komposer" and "Kontext-powered Presets
Now you can upload your image and ask to change it by simply selecting a preset, rather than writing out a complex prompt. A one-button solution for those who don't know how\don't want to make prompts.
It's clear that under the hood there is some kind of system prompt for each preset (which LLM I don't know). So smart people first dug up all these prompts, then screwed them into Komfi and even tried them in other generators\chats.
🔥521❤517👍497🎉487
Here we have another competitor to Flux Kontext
This time from Nvidia.
This is an inserter of objects into pictures.
You can see sweet examples here:
https://research.nvidia.com/labs/par/addit/
This time from Nvidia.
This is an inserter of objects into pictures.
You can see sweet examples here:
https://research.nvidia.com/labs/par/addit/
👍57🔥53❤48🎉44
Everyone has probably already read that Grok has brought in a couple of avatars, which they call companions.
These are Girl Ani and red panda Bad Ruby, they say that there will be a third and fourth.
So far on a SuperGrok subscription and in the iOS app.
You can turn on NSFW mode in their Settings and then they start to burn in full virtual power.
Look how brilliantly Deemos Tech, the developers of one of the best 3D generators Rodin, responded.
They simply took this very Ani and converted it into 3D, then did some magic with the material and shoved it into Mixamo for rigging and animation. With Ruby it turned out even more elegant.
And now Ani and Ruby are dancing to the user's tune.
A great example of how one hyped product makes great advertising based on another hyped product.
These are Girl Ani and red panda Bad Ruby, they say that there will be a third and fourth.
So far on a SuperGrok subscription and in the iOS app.
You can turn on NSFW mode in their Settings and then they start to burn in full virtual power.
Look how brilliantly Deemos Tech, the developers of one of the best 3D generators Rodin, responded.
They simply took this very Ani and converted it into 3D, then did some magic with the material and shoved it into Mixamo for rigging and animation. With Ruby it turned out even more elegant.
And now Ani and Ruby are dancing to the user's tune.
A great example of how one hyped product makes great advertising based on another hyped product.
🎉580❤575👍563🔥561
Media is too big
VIEW IN TELEGRAM
High Input Fidelity for chatGPT
Precise image editing in GPT Image-1.
Until now it was clear that, unlike Flux Kontext, chatGPT adds quite a lot of detail to faces, poses, and environments when editing images.
They have just announced the High Input Fidelity parameter, which is designed to preserve details and more accurately preserve the appearance of the original image.
It works only through API and only on the platform https://platform.openai.com/playground/images/
Precise image editing in GPT Image-1.
Until now it was clear that, unlike Flux Kontext, chatGPT adds quite a lot of detail to faces, poses, and environments when editing images.
They have just announced the High Input Fidelity parameter, which is designed to preserve details and more accurately preserve the appearance of the original image.
It works only through API and only on the platform https://platform.openai.com/playground/images/
❤96👍86🔥85🎉80
This media is not supported in your browser
VIEW IN TELEGRAM
Suno v4.5+
More variety in genres
Thicker vocals
Better prompt understanding
Speed up generation time
More creative beats: Specific improvements for genres like metal and pop-punk aimed at reducing repetition and increasing variety.
More complex harmonic arrangements
Stronger Covers support: can imitate celebrity voices and simulate specific roles and styles
Add Vocals - can be added to generations or to external files. Resko speeds up vocal selection.
Add Instrumentals does the opposite: you provide a vocal file (voice recording or vocal fragment), and the AI can immediately generate basic accompaniment for it to create a full song. The same vocals can be quickly used in different instrumental styles.
More variety in genres
Thicker vocals
Better prompt understanding
Speed up generation time
More creative beats: Specific improvements for genres like metal and pop-punk aimed at reducing repetition and increasing variety.
More complex harmonic arrangements
Stronger Covers support: can imitate celebrity voices and simulate specific roles and styles
Add Vocals - can be added to generations or to external files. Resko speeds up vocal selection.
Add Instrumentals does the opposite: you provide a vocal file (voice recording or vocal fragment), and the AI can immediately generate basic accompaniment for it to create a full song. The same vocals can be quickly used in different instrumental styles.
🔥176👍164❤157🎉154
ChatGPT has introduced restyling at the button level, not at the prompt level. The "styles" are just pre-made prompts.
❤396🔥366👍357🎉340
This media is not supported in your browser
VIEW IN TELEGRAM
Looks like Musk, with his avatars Anya and Rudy, has pulled the emergency brake.
Hedra has just dropped real-time avatars.
They're definitely stepping into the territory of HeyGen and Character ai here, but it seems we're in for real-time avatars that can be connected not just to Grok or custom models, but to any LLM at all.
At least, that’s how it looks with Hedra.
Hedra has just dropped real-time avatars.
They're definitely stepping into the territory of HeyGen and Character ai here, but it seems we're in for real-time avatars that can be connected not just to Grok or custom models, but to any LLM at all.
At least, that’s how it looks with Hedra.
🎉275❤264🔥261👍260
🚨 OpenAI is preparing to launch GPT-5 in August.
(OpenAI is also trying to release its open-source LRM by the end of July. According to a source, it will be comparable to o3-mini.)
UPD: There will also be a GPT-5-mini, available in ChatGPT and the API, and a nano version, available only through the API.
UPD 2: The current plan is to launch GPT-5 in early August, but that timeline might shift.
(OpenAI is also trying to release its open-source LRM by the end of July. According to a source, it will be comparable to o3-mini.)
UPD: There will also be a GPT-5-mini, available in ChatGPT and the API, and a nano version, available only through the API.
UPD 2: The current plan is to launch GPT-5 in early August, but that timeline might shift.
👍136🔥126🎉120❤117
This media is not supported in your browser
VIEW IN TELEGRAM
Higgsfield Steal
Steal allows users to recreate any image from the internet using AI. The user simply selects an image they like, and the system instantly transfers its aesthetics—such as clothing, pose, lighting, and atmosphere—onto another image.
Steal allows users to recreate any image from the internet using AI. The user simply selects an image they like, and the system instantly transfers its aesthetics—such as clothing, pose, lighting, and atmosphere—onto another image.
👍186🎉177❤175🔥173
Media is too big
VIEW IN TELEGRAM
Kling: they introduced Kling Lab — an infinite canvas for creating different versions of content and storylines. It has a built-in chat for generating and editing text, images, and video via prompts. Collaborative work is supported.
Similar things are being developed by Flora, as well as platforms like Runway and Midjourney.
It's still in beta. If anyone has access, share your impressions in the comments.
Similar things are being developed by Flora, as well as platforms like Runway and Midjourney.
It's still in beta. If anyone has access, share your impressions in the comments.
🎉224👍211❤207🔥199
This media is not supported in your browser
VIEW IN TELEGRAM
Suno now has its own radio.
There are periodic votes on "what are we going to stream?"
There are periodic votes on "what are we going to stream?"
👍225🔥219❤218🎉216
OpenAI Introduced study mode in ChatGPT — step by step guidance for students rather than quick answers.
Openai
Introducing study mode
Introducing study mode in ChatGPT, a new learning experience that helps you work through problems step by step, guiding students with questions, scaffolding, and feedback for deeper learning.
👍175❤167🔥152🎉136
New FEATURE – Ideogram Character: just one photo, and then the head flies into the Latent with IP adapters and other internal stuff. By the way, it handles head rotations pretty well too. Overall, it turns out really cool.
P.S. From now on, I'm going to dress like in the picture.
P.S. From now on, I'm going to dress like in the picture.
👍514❤511🎉507🔥477
Forwarded from BlockChainWORLD.ai - daily crypto and AI news and promos!
This media is not supported in your browser
VIEW IN TELEGRAM
Missing out on the AI profits? Make your own viral faceless videos or ads in seconds with simple templates!
I’ve tested over 500 tools, and only ONE replaced my entire video workflow — Vortex AI.
Viral templates, cinematic effects, top models like Veo3, image and 3D generation — all in one platform.
Just type your idea, pick a style, and generate viral content in minutes.
Go to https://vortex.channel/ and get 100 free credits.
I’ve tested over 500 tools, and only ONE replaced my entire video workflow — Vortex AI.
Viral templates, cinematic effects, top models like Veo3, image and 3D generation — all in one platform.
Just type your idea, pick a style, and generate viral content in minutes.
Go to https://vortex.channel/ and get 100 free credits.
🔥170🎉162👍157❤145
Luma Modify with Instructions in Dream Machine
Luma released something that’s kind of similar to Runway Aleph.
Also video editing and neural compositing.
But I dove into Twitter. People there are saying that Luma doesn’t hold a candle to Runway.
And Runway Aleph, by the way, was given out to paid subscribers.
Luma released something that’s kind of similar to Runway Aleph.
Also video editing and neural compositing.
But I dove into Twitter. People there are saying that Luma doesn’t hold a candle to Runway.
And Runway Aleph, by the way, was given out to paid subscribers.
🎉358🔥355❤348👍319