Midjourney v6.1 & Leonardo.AI Acquisition!
TLDRThis week saw the launch of Midjourney's v6.1 AI image generation model, promising sharper images and improved text rendering. The update introduces new features like personalized image enhancement and texture-rich Q mode. Meanwhile, Canva's acquisition of Leonardo.AI could impact Adobe's dominance with its creative AI technology. Flux, an open-source alternative to Midjourney, was also released. Runway ml's Gen 3 pricing is set to become more accessible, with a faster 'turbo' model on the horizon.
Takeaways
- 🌟 Midjourney has released its v6.1 model, which is set as the default and offers sharper image quality, more coherent outputs, improved text rendering, and an enhanced upscaler.
- 🆕 The v6.1 model shows noticeable improvements in image generation, especially with personalization codes added to prompts, although the changes are not as dramatic as between v5 and 5.1.
- 🔍 A new 'Q' mode has been introduced that increases image texture, potentially at the cost of coherence, offering a creative but possibly less coherent image output.
- 📈 The upscaler in v6.1 is recommended to use the 'subtle' setting for most creative work, as the 'creative' setting can be too heavy-handed and airbrushed.
- 📝 Enhanced in-image text coherence is a feature of v6.1, where text within quotation marks is rendered more accurately.
- 🎨 Midjourney's image reference feature is updated to take more details from the initial image and integrate them into the rendering.
- 🔮 Upcoming version 7 of Midjourney promises enhanced aesthetics, faster performance, smarter prompt understanding, and significant overall enhancements, with 3D and video still on the roadmap.
- 🛍️ Canva has acquired Leonardo.ai, which will continue to operate independently, potentially integrating its Phoenix model into Canva's Magic Media feature.
- 🎨 Flux, an open-source text-to-image model by Black Forest Labs, positions itself as a competitor to Midjourney and is worth exploring for its capabilities.
- 💰 Runway ML's Gen 3 pricing is being adjusted to offer a new 'turbo' model for faster video generation with significantly lower pricing, addressing previous cost concerns.
- 🔄 The integration of Leonardo into Canva and Affinity products could potentially offer a more creative alternative to Adobe's Gen AI, shaking up the industry.
Q & A
What significant update did Midjourney release in its AI image generation model?
-Midjourney released its v6.1 model, which is touted to have sharper image quality, more coherent outputs, improved text rendering, and an enhanced upscaler.
What is the difference between Midjourney's v6.1 and previous versions in terms of image quality?
-The v6.1 update provides subtle improvements in image quality, such as sharper images and better text rendering, but the differences are not as dramatic as the jump from version 5 to 5.1.
How can users customize their AI-generated images with Midjourney's v6.1 model?
-Users can customize their images by adding a personalization code or by appending '-d-p' to the end of their prompt, which will allow the model to generate images with improved nuance and accuracy.
What is the Q mode in Midjourney's v6.1 model, and how does it affect the images?
-The Q mode, activated with the command '--Q Space 2', increases the textures of the image but might reduce the coherence of the image to some extent.
What advice is given regarding the use of upscalers in Midjourney's v6.1 model?
-It is suggested that users should lean more on the 'subtle' upscaler for most creative work, as the 'creative' upscaler may be too heavy-handed and give an airbrushed effect.
What is the new feature in Midjourney's v6.1 model that improves text coherence within images?
-The new feature allows for better in-image text coherence when using quotation marks with a word, ensuring that the word appears as intended within the generated image.
What is the significance of Canva's acquisition of Leonardo.AI?
-The acquisition of Leonardo.AI by Canva is significant as it suggests a potential integration of Leonardo's creative AI technology into Canva's suite of design tools, possibly enhancing their capabilities in image generation and editing.
What does the future hold for Midjourney's AI image generation model after version 6.1?
-After version 6.1, Midjourney plans to release version 6.2 within a month, which will be followed by version 7. The key features of version 7 are expected to include enhanced aesthetics, faster performance, smarter prompt understanding, and significant overall enhancements.
What is the relationship between Canva's acquisition of Affinity and Leonardo.AI?
-Canva's acquisition of both Affinity and Leonardo.AI could potentially lead to the integration of Leonardo's AI technology into Affinity's photo editing software, offering a more creative and advanced alternative to Adobe's Photoshop.
What is the update on Runway ML's Gen 3 pricing, and how does it affect users?
-Runway ML is planning to roll out a turbo model for Gen 3 that will generate videos much faster. Additionally, they are working on significantly lower pricing for the image-to-video feature and plan to make it available to free users, addressing previous concerns about cost.
Outlines
🚀 Mid Journey v6.1 Model Release and Features
The script discusses the launch of Mid Journey's v6.1 model, which promises improved image quality, coherence, text rendering, and an enhanced upscaler. The comparison is made with previous versions, highlighting the subtle yet noticeable improvements in image generation. The script also covers personalization features, such as adding '-d' to prompts for more nuanced results, and the introduction of a new Q mode for texture enhancement. Additionally, the discussion includes the use of the upscaler and the potential impact on text coherence within images.
🌐 Updates on AI Imagery and Industry Acquisitions
This paragraph covers several updates in the AI imagery space. It mentions the acquisition of Leonardo by Canva, which is significant given Canva's recent acquisition of Affinity, a Photoshop alternative. The script speculates on the potential integration of Leonardo's Phoenix model into Canva's Magic Media feature and the implications for Adobe's Gen Phil. It also introduces Flux, an open-source text-to-image model by Black Forest Labs, as a potential competitor to Mid Journey. The paragraph concludes with a discussion on Runway ML's Gen 3 pricing, indicating that a more affordable turbo model for image-to-video is on the horizon.
📉 Response to Runway Gen 3 Pricing Concerns
The final paragraph addresses the community's concerns regarding the cost of using Runway Gen 3. It acknowledges the high prices as a significant issue and commends Runway for introducing a new turbo model that promises faster video generation at a lower cost. The script clarifies that the previously mentioned $95 unlimited plan was not the new pricing structure and that an official announcement on pricing has not yet been made. It ends with the anticipation of Runway's final pricing decision and a note of thanks to the viewers.
Mindmap
Keywords
Midjourney v6.1
AI Image Generation
Open Source
Leonardo.AI Acquisition
Runway ML
Upscale
Personalization Code
Q Mode
In-Image Text Coherence
Describe
Adobe
Highlights
Midjourney v6.1 model release brings improved image quality, coherence, text rendering, and upscaler.
Comparison between v6.1 and previous versions shows subtle yet notable enhancements in image generation.
Personalization code addition to prompts refines the output nuances and accuracy in v6.1.
New 'Q mode' increases image texture but may affect coherence.
Upscale options in v6.1, with 'subtle' recommended for most creative use.
In-image text coherence improved with quotation marks in v6.1.
Runway ML's Gen 3 pricing remains unchanged, contrary to initial rumors.
Flux, an open-source text-to-image model by Black Forest Labs, positions as a mid-journey competitor.
Canva's acquisition of Leonardo raises questions about potential integration with Canva's Magic Media feature.
Leonardo's independence post-acquisition ensures continued innovation in AI imagery.
Canva's recent acquisition of Affinity Photo hints at a possible integration with Leonardo's technology.
Affinity Photo, built by ex-Photoshop developers, offers a one-time payment alternative to Adobe's subscription model.
Runway's Gen 3 Turbo promises faster video generation with lower pricing for users.
Runway addresses user concerns regarding cost with upcoming adjustments to Gen 3 pricing.
The upcoming version 7 of Midjourney promises significant enhancements in aesthetics, performance, and rendering.
3D and video capabilities, along with the Storyteller tool, are on Midjourney's development roadmap.
Describe feature in Midjourney is currently undergoing updates, improving image reference integration.