BREAKING NEWS
LATEST POSTS
-
Seaweed APT – Diffusion Adversarial Post-Training for One-Step Video Generation
https://cdn.seaweed-apt.com/assets/showreel/seaweed-apt.mp4
This demonstrate large-scale text-to-video generation with a single neural function evaluation (1NFE) by using our proposed adversarial post-training technique. Our model generates 2 seconds of 1280×720 24fps videos in real-time

-
Pyper – a flexible framework for concurrent and parallel data-processing in Python
Pyper is a flexible framework for concurrent and parallel data-processing, based on functional programming patterns.
https://github.com/pyper-dev/pyper

-
Jacob Bartlett – Apple is Killing Swift
https://blog.jacobstechtavern.com/p/apple-is-killing-swift
Jacob Bartlett argues that Swift, once envisioned as a simple and composable programming language by its creator Chris Lattner, has become overly complex due to Apple’s governance. Bartlett highlights that Swift now contains 217 reserved keywords, deviating from its original goal of simplicity. He contrasts Swift’s governance model, where Apple serves as the project lead and arbiter, with other languages like Python and Rust, which have more community-driven or balanced governance structures. Bartlett suggests that Apple’s control has led to Swift’s current state, moving away from Lattner’s initial vision.

-
Don’t Splat your Gaussians – Volumetric Ray-Traced Primitives for Modeling and Rendering Scattering and Emissive Media
https://arcanous98.github.io/projectPages/gaussianVolumes.html
We propose a compact and efficient alternative to existing volumetric representations for rendering such as voxel grids.

-
IPAdapter – Text Compatible Image Prompt Adapter for Text-to-Image Image-to-Image Diffusion Models and ComfyUI implementation
github.com/tencent-ailab/IP-Adapter
The IPAdapter are very powerful models for image-to-image conditioning. The subject or even just the style of the reference image(s) can be easily transferred to a generation. Think of it as a 1-image lora. They are an effective and lightweight adapter to achieve image prompt capability for the pre-trained text-to-image diffusion models. An IP-Adapter with only 22M parameters can achieve comparable or even better performance to a fine-tuned image prompt model.
Once the IP-Adapter is trained, it can be directly reusable on custom models fine-tuned from the same base model.The IP-Adapter is fully compatible with existing controllable tools, e.g., ControlNet and T2I-Adapter.


-
SPAR3D – Stable Point-Aware Reconstruction of 3D Objects from Single Images
SPAR3D is a fast single-image 3D reconstructor with intermediate point cloud generation, which allows for interactive user edits and achieves state-of-the-art performance.
https://github.com/Stability-AI/stable-point-aware-3d
https://stability.ai/news/stable-point-aware-3d?utm_source=x&utm_medium=social&utm_campaign=SPAR3D
-
MiniMax-01 goes open source
MiniMax is thrilled to announce the release of the MiniMax-01 series, featuring two groundbreaking models:
MiniMax-Text-01: A foundational language model.
MiniMax-VL-01: A visual multi-modal model.Both models are now open-source, paving the way for innovation and accessibility in AI development!
🔑 Key Innovations
1. Lightning Attention Architecture: Combines 7/8 Lightning Attention with 1/8 Softmax Attention, delivering unparalleled performance.
2. Massive Scale with MoE (Mixture of Experts): 456B parameters with 32 experts and 45.9B activated parameters.
3. 4M-Token Context Window: Processes up to 4 million tokens, 20–32x the capacity of leading models, redefining what’s possible in long-context AI applications.💡 Why MiniMax-01 Matters
1. Innovative Architecture for Top-Tier Performance
The MiniMax-01 series introduces the Lightning Attention mechanism, a bold alternative to traditional Transformer architectures, delivering unmatched efficiency and scalability.2. 4M Ultra-Long Context: Ushering in the AI Agent Era
With the ability to handle 4 million tokens, MiniMax-01 is designed to lead the next wave of agent-based applications, where extended context handling and sustained memory are critical.3. Unbeatable Cost-Effectiveness
Through proprietary architectural innovations and infrastructure optimization, we’re offering the most competitive pricing in the industry:
$0.2 per million input tokens
$1.1 per million output tokens🌟 Experience the Future of AI Today
We believe MiniMax-01 is poised to transform AI applications across industries. Whether you’re building next-gen AI agents, tackling ultra-long context tasks, or exploring new frontiers in AI, MiniMax-01 is here to empower your vision.✅ Try it now for free: hailuo.ai
📄 Read the technical paper: filecdn.minimax.chat/_Arxiv_MiniMax_01_Report.pdf
🌐 Learn more: minimaxi.com/en/news/minimax-01-series-2
💡API Platform: intl.minimaxi.com/

FEATURED POSTS
-
Photography basics: Shutter angle and shutter speed and motion blur
http://www.shutterangle.com/2012/cinematic-look-frame-rate-shutter-speed/
https://www.cinema5d.com/global-vs-rolling-shutter
https://www.wikihow.com/Choose-a-Camera-Shutter-Speed
Shutter is the device that controls the amount of light through a lens. Basically in general it controls the amount of time a film is exposed.
Shutter speed is how long this device is open for, which also defines motion blur… the longer it stays open the blurrier the image captured.
The number refers to the amount of light actually allowed through.As a reference, shooting at 24fps, at 180 shutter angle or 1/48th of shutter speed (0.0208 exposure time) will produce motion blur which is similar to what we perceive at naked eye
Talked of as in (shutter) angles, for historical reasons, as the original exposure mechanism was controlled through a pie shaped mirror in front of the lens.

A shutter of 180 degrees is blocking/allowing light for half circle. (half blocked, half open). 270 degrees is one quarter pie shaped, which would allow for a higher exposure time (3 quarter pie open, vs one quarter closed) 90 degrees is three quarter pie shaped, which would allow for a lower exposure (one quarter open, three quarters closed)
(more…)
-
The Forbidden colors – Red-Green & Blue-Yellow: The Stunning Colors You Can’t See
www.livescience.com/17948-red-green-blue-yellow-stunning-colors.html

While the human eye has red, green, and blue-sensing cones, those cones are cross-wired in the retina to produce a luminance channel plus a red-green and a blue-yellow channel, and it’s data in that color space (known technically as “LAB”) that goes to the brain. That’s why we can’t perceive a reddish-green or a yellowish-blue, whereas such colors can be represented in the RGB color space used by digital cameras.
https://en.rockcontent.com/blog/the-use-of-yellow-in-data-design
The back of the retina is covered in light-sensitive neurons known as cone cells and rod cells. There are three types of cone cells, each sensitive to different ranges of light. These ranges overlap, but for convenience the cones are referred to as blue (short-wavelength), green (medium-wavelength), and red (long-wavelength). The rod cells are primarily used in low-light situations, so we’ll ignore those for now.
When light enters the eye and hits the cone cells, the cones get excited and send signals to the brain through the visual cortex. Different wavelengths of light excite different combinations of cones to varying levels, which generates our perception of color. You can see that the red cones are most sensitive to light, and the blue cones are least sensitive. The sensitivity of green and red cones overlaps for most of the visible spectrum.

Here’s how your brain takes the signals of light intensity from the cones and turns it into color information. To see red or green, your brain finds the difference between the levels of excitement in your red and green cones. This is the red-green channel.
To get “brightness,” your brain combines the excitement of your red and green cones. This creates the luminance, or black-white, channel. To see yellow or blue, your brain then finds the difference between this luminance signal and the excitement of your blue cones. This is the yellow-blue channel.
From the calculations made in the brain along those three channels, we get four basic colors: blue, green, yellow, and red. Seeing blue is what you experience when low-wavelength light excites the blue cones more than the green and red.
Seeing green happens when light excites the green cones more than the red cones. Seeing red happens when only the red cones are excited by high-wavelength light.
Here’s where it gets interesting. Seeing yellow is what happens when BOTH the green AND red cones are highly excited near their peak sensitivity. This is the biggest collective excitement that your cones ever have, aside from seeing pure white.
Notice that yellow occurs at peak intensity in the graph to the right. Further, the lens and cornea of the eye happen to block shorter wavelengths, reducing sensitivity to blue and violet light.









