Stability AI just rolled out Japanese InstructBLIP Alpha, its first Japanese vision-language model. Built on top of its 7B Japanese LLM and the InstructBLIP architecture, it handles image captioning and visual QA while recognizing Japan-specific landmarks like Tokyo Skytree. It is currently available on Hugging Face for research purposes.
Stability AI just partnered with NVIDIA to optimize SDXL using TensorRT, releasing open ONNX weights globally. The integration doubles performance on NVIDIA H100 GPUs, spitting out HD images in 1.47 seconds with a 41% latency drop and 70% throughput jump. Why it matters: high-end open-source image generation just got significantly faster and cheaper to scale.
Stability AI just debuted Japanese StableLM Alpha, a 7B-parameter model marking its entrance into the Japanese language space. Trained on 750 billion tokens, the base model is available for commercial use under Apache 2.0, while the instruction-tuned version tops public benchmarks for research.
Stability AI just launched Stable Code Alpha 3B, its first generative AI model tailored for coding. Trained on 560B tokens of code, the suite includes base, instruction, and long-context variants with a 16,000-token window capable of reviewing multiple files at once. Why it matters: It gives developers a lightweight, accessible open-source assistant for everyday programming.
AI creators just revived digital artist Greg Rutkowski's style using a custom LoRA model on Civitai, bypassing Stability AI's official removal. Rutkowski had campaigned to purge his work from Stable Diffusion 2.0, but open-source creators stepped in to recreate his aesthetic anyway. Why it matters: It shows that in an open-source world, opt-out requests are practically impossible to enforce.
Stability AI just open-sourced SDXL 1.0, its flagship text-to-image model. It outputs native 1024x1024 images, handles notoriously tricky hands and text much better, and ditches the need for prompt bloat like "masterpiece." A two-stage 6.6B pipeline runs on consumer GPUs with just 8GB of VRAM.
InvokeAI just launched version 3.0.0, overhauling its codebase around an extensible node architecture with a new visual node editor. The release brings built-in ControlNet controls, preliminary SDXL support, long prompts exceeding 77 tokens, and half-precision loading on Mac to slash RAM usage.
A US federal judge just signaled he is inclined to toss most claims in an artist-led copyright lawsuit against Stability AI, Midjourney, and DeviantArt. Judge William Orrick ruled that the artists failed to show substantial similarity between their works and AI outputs, though he gave them a chance to file a revised complaint with stronger evidence.
Stability AI just rolled out a revamped developer platform that combines interactive API documentation with a built-in sandbox. Developers can now tweak prompts, test image-to-image or upscaling workflows in real time, and manage API keys directly without jumping through DreamStudio.
Stability AI just released Stable Beluga 1 and 2, two instruction-tuned language models built on Meta's LLaMA and LLaMA 2. Inspired by Microsoft's Orca paper, the team used just 600,000 synthetic data points—roughly 10% of Orca's dataset size—to climb to the number one spot on Hugging Face's Open LLM Leaderboard. Both models are available for download under a non-commercial research license.
Stability AI CEO Emad Mostaque just warned that most outsourced coders in India could lose their jobs within the next year or two as generative AI drastically shrinks the headcount needed for software development. He pointed out that back-office roles away from human view are prime targets, though AI copilots will also push engineers toward higher-level collaboration.
Leonardo AI just integrated Stability AI's SDXL 0.9 model alongside Prompt Magic v3, sweetening the rollout with 2,000 bonus token credits. The dual-stage pipeline supports native 1024x1024 resolution and includes a refinement module that boosts photorealism and lighting accuracy. It gives creators a noticeable bump in fidelity without needing to switch out of Leonardo's workflow.
Stability AI's Clipdrop just launched Stable Doodle, a sketch-to-image tool powered by SDXL 0.9 and Tencent ARC's T2I-Adapter. Anyone can sketch a rough outline, type a prompt, and pick from 14 styles to generate high-quality images in seconds. It bridges the gap between raw imagination and detailed prompts, and it is free to try online with daily limits.
Stability AI just launched SDXL 0.9, bringing a significant upgrade to its flagship text-to-image pipeline over the April beta. It uses a 6.6B parameter two-stage ensemble to generate high-res 1024x1024 visuals while still fitting on a consumer GPU with 8GB VRAM. Why it matters: High-fidelity open models are closing the gap with closed competitors, with a full open release slated for mid-July.
Clipdrop just released Uncrop, an outpainting tool built on Stable Diffusion XL that extends image backgrounds to complement existing photos and adjust aspect ratios. Users can resize images or fix framing without writing prompts, and the tool is available for free on the web app without requiring a login.
Stability AI's Clipdrop just launched Reimagine XL, an upgraded visual tool that creates brand-new variations of any uploaded photo without touching its original pixels. Instead of relying on text prompts, it uses an advanced image encoder to output high-resolution alternatives across diverse aspect ratios.
Stability AI just open-sourced the frontend behind DreamStudio, launching it as a community-driven project called StableStudio. The move hands the keys of its primary image generation interface over to external devs, with roadmaps for WebGPU local inference, plugins, and an upcoming chat tool. Stability will keep DreamStudio alive as a hosted implementation of the codebase.
Stability AI just launched the Stable Animation SDK, giving artists and developers programmatic tools to generate AI animations using Stable Diffusion models. Users can create animations through text prompts alone, source images, or existing videos. The release makes it much easier to plug diffusion-based animation straight into developer pipelines.
Google just announced the integration of WebGPU into Chrome at Google I/O, enabling web apps to tap directly into laptop and phone GPUs. The company also demoed a web-based Stable Diffusion tool generating images directly on-device. Why it matters: Shifting AI from massive cloud servers to local browsers slashes latency while keeping private data on your own machine.
Stability AI just released StableVicuna-13B, billed as the first large-scale open-source chatbot fine-tuned using reinforcement learning from human feedback (RLHF). Built on Vicuna-13B and trained using trlX alongside open preference datasets from Anthropic and OpenAssistant, it brings the post-training playbook behind ChatGPT into open source. Weights are available on Hugging Face as a delta over original LLaMA weights.
