ComfyUI – v0.22.2

ComfyUI v0.22.2 is here! 🚀

If you live for node-based workflows and complex Stable Diffusion pipelines, it’s time to check your updates. ComfyUI continues to be the go-to modular engine for anyone wanting to push the limits of SDXL, ControlNet, and LCM models without writing a single line of code.

While the specific changelog details are currently being processed, this update arrives as part of the ongoing evolution of the platform’s decoupled architecture. The frontend is now hosted in its own separate repository, which is a huge win for us tinkerers because it means much more frequent UI updates and smoother performance tweaks without waiting on backend overhauls. 🛠️

What to keep an eye on with this version:

• Enhanced Frontend Agility: Expect faster iterations on the visual interface and node connection logic.
• Optimized Workflow Execution: Continued improvements to smart memory management for those heavy-duty upscaling and video generation tasks.
• Seamless Integration: Better stability for custom nodes and multi-GPU setups.

Keep those workflows flowing! If you’re running heavy SDXL or LoRA-heavy pipelines, a quick update check is always a good idea to ensure your node connections stay rock solid. 🎨✨

🔗 View Release

Ollama – v0.30.0-rc23

Ollama just dropped v0.30.0-rc23! 🛠️

If you aren’t using Ollama yet, it is the absolute gold standard for running powerful LLMs like Llama 3, DeepSeek-R1, and Mistral locally on your own hardware. It takes all the headache out of model management with a super slick CLI and a built-in REST API that makes integrating local AI into your own dev projects a breeze.

This latest release candidate is all about stability and polishing the foundation. While it’s not a massive feature overhaul, these maintenance tweaks are crucial for a reliable local environment:

• Code Cleanup: The team focused heavily on lint fixes, meaning cleaner, more standardized code that helps squash bugs before they reach your machine.
• Improved Build Process: Updates to asset management ensure smoother deployments and more consistent builds.

It’s a “under the hood” kind of update designed to keep your local inference running smoothly. Keep those engines idling—the full stable v0.30.0 release is right around the corner! 🚀

🔗 View Release

ComfyUI – v0.22.1

ComfyUI v0.22.1 is officially live! 🚀

If you’re deep in the world of node-based generative art, you know ComfyUI is the ultimate playground for building complex Stable Diffusion pipelines. Whether you’re chaining ControlNets or fine-tuning SDXL workflows, this update keeps your engine running smoothly.

This latest patch (v0.22.1) focuses on tightening up the core experience:

• Bug Squashing: Fixed those annoying little glitches that can disrupt complex node executions mid-workflow.
• Enhanced Stability: Improved reliability to prevent unexpected crashes during heavy-duty rendering sessions.
• Performance Polishing: Refined underlying code logic to ensure your generations stay snappy and efficient. 🛠️

Pro-tip for the tinkerers: Since this is a patch release, it’s a great time to double-check that your favorite custom nodes are up to date to maintain full compatibility with the core engine!

🔗 View Release

Tater – Tater v81

🥔 Tater v81 — “The Potato Learned Terminal Commands” 💻

Hold onto your hats, tinkerers! Tater just leveled up from a simple chat app to a full-blown local AI workspace. This update is massive—the potato officially has shell access now, meaning it can actually execute tasks instead of just talking about them.

Here’s the breakdown of what’s new:

• Spudex Terminal Workbench: A brand new main tab that acts as a full terminal-console workspace. You can chat with terminal-capable models, run commands, write scripts, and inspect outputs directly within Tater. It’s basically a coding environment powered by AI. 🛠️
• Hydra Terminal Tools: Hydra is no more “pretending” to do work. With the new `run_terminal_task` tool, it can handle real multi-step terminal tasks and will even auto-continue unfinished work until the job is done.
• Enhanced Guardrails: Since giving an AI shell access is… well, risky… Spudex includes built-in controls. You can set command restrictions, manage network/install permissions, track processes, and require approvals before any “dangerous” commands run. Use containers to stay safe! 🛡️

Self-Configuring Wake Words: Using the new Firmware v3.0.3, Tater can now change its own satellite wake words via voice command (e.g., “Tater, change your wake word to hey computer”*). It can even suggest close matches or reset to defaults. 🎙️

• Smarter Web Research: Hydra’s web research has moved past “AI confidence theater.” It can now actively search, inspect pages, and evaluate if it actually found the answer, continuing its research loop if the initial results are insufficient. 🌐
• Personal Core Dashboard: Your dashboard just got a lot more useful with integrated 7-day outlook context, multi-user selection, and calendar/subscription briefs.

Tater is evolving into a persistent runtime system for your local workflows. Just remember: use common sense (and maybe an isolated folder) when letting the potato run scripts! 🥔⚠️

🔗 View Release

Tater – Tater v80

🥔 Tater v80 — “Wake Word Wars” is here! 😏🎙️

Get ready to turn your local AI setup into a full-blown voice command laboratory! Tater is breaking free from single-engine dependency, allowing you to mix and match different detection engines live from your firmware entities. It’s all about modularity and making your voice ecosystem as flexible as possible.

Here’s the scoop on v80:

• The NanoWakeWord Expansion: Your Tater satellites now support `microWakeWord`, `openWakeWord`, and the brand new `NanoWakeWord` all within the same ecosystem. 🛰️
• Dedicated NanoWakeWord Server: Heavy lifting? Not anymore! You can now offload processing to a standalone server with Docker and NVIDIA Docker support. It features dedicated WebSocket wake streaming and precise score handling, keeping your main Tater instance lightweight and snappy.
• NanoWakeWord Trainer: For the true devs and tinkerers, a dedicated training flow is officially live! 🏋️ You can now handle synthetic sample generation, validation, and seamless ONNX + PyTorch model syncing to bring your own custom-trained models into the family.
• openWakeWord Server Cleanup: We’ve silenced the noise! Diagnostic detection logging is now optional and disabled by default, so your logs will finally stop screaming at you. 🧠

Tater is evolving from a single engine into a massive collection of tools. Whether you want to experiment with local vs. remote detection or deploy custom-trained models, v80 gives you the keys to the kingdom! 🛠️

🔗 View Release

Lemonade – v10.6.0: fix: emit collection components under public model names (#1985)

🍋 Lemonade SDK v10.6.0 is here! 🍋

If you’ve been running LLMs locally using Lemonade to squeeze every bit of power out of your NPU or GPU, listen up! This update is a massive win for anyone building custom workflows or managing model components via the SDK. It smooths out some nasty synchronization issues that were causing headaches in the UI.

What’s new in this release:

• Fixed Component Naming Mismatch: No more “not downloaded” errors! Previously, collection components were being serialized with an internal `user.` prefix, which broke renderers looking for the public names. Now, the SDK correctly emits components under their public model names, keeping your wire format consistent and your download buttons working perfectly. 🛠️
• Improved Reliability: The fix ensures that the Planner LLM field no longer renders as a blank space when you’re dealing with user-registered components.
• Robust Testing: The devs have updated the test suites to ensure this new public-name format works flawlessly without messing up internal cache lookups or storage canonicalization.

The best part? Your load/download paths and cache keys remain exactly the same, so upgrading is a total breeze for your dev workflow! 🚀

🔗 View Release

Home Assistant Voice Pe – 26.5.0

Home Assistant Voice PE 26.5.0 is live! 🚀

If you’ve been building your own local voice assistant using ESPHome, this update is a massive win for stability and custom pipelines. This project is all about bringing private, offline voice control to your smart home, and this patch makes those interactions much more seamless.

What’s new in this release:

• Enhanced Stability & Sync: Say goodbye to timing hiccups! This update focuses on improved Sendspin stability and significantly better sync accuracy to keep your audio streams aligned.
• Home Assistant Integration Boost: A huge win for automation nerds! Home Assistant can now request unprocessed audio specifically for STT (Speech-to-Text) engines that handle their own processing internally. This makes custom voice pipelines much smoother and more efficient.

Big shout out to @teancommade for making their first contribution to the project! 🛠️

🔗 View Release

Ollama – v0.30.0-rc22

Ollama v0.30.0-rc22 🛠️

If you’re obsessed with running powerful LLMs like Llama 3 or DeepSeek-R1 locally without the cloud headache, Ollama is your best friend. It’s the ultimate toolkit for downloading, managing, and chatting with open-source models right on your own hardware.

This latest release candidate update is all about polishing the experience as we approach the official v0.30.0 stable launch!

What’s new:

• Refinement Step: This version (v0.30.0-rc22) serves as a critical bridge in the pre-release cycle, focusing on fine-tuning the runner performance.
• Stability Focus: As an RC update, the primary goal here is squashing bugs and ensuring everything is rock-solid before the big stable rollout hits your machine.

Keep those engines idling—these frequent RC bumps mean the heavy-hitting new features are just around the corner! 🚀

🔗 View Release

Tater – Tater v81

🥔 Tater v81 — “The Potato Learned Terminal Commands” 💻

Hold onto your hats, tinkerers! Tater just leveled up from a local chat app to a full-blown local AI workspace and runtime system. The potato officially has shell access now, so things are about to get much more interesting (and potentially chaotic)! ⚠️

🚀 Spudex: The New Terminal Workbench

The biggest highlight is Spudex, a brand-new terminal-console workspace built directly into Tater. It’s essentially a “Codex” for your local machine where you can:

• Chat with terminal-capable models.
• Run commands, write scripts, and manage files.
• Inspect outputs and retry/fix failed tasks in real-time.

Pro-tip:* Use containers or isolated folders! Spudex is powerful, so keep those guardrails active to prevent the AI from “reorganizing” your filesystem into abstract art. 🎨

🧠 Hydra Terminal Tools & Smarter Research

Hydra is getting much more capable with new tools like `spudex_run`, `spudex_task`, and `spudex_status`.

• Multi-step execution: Hydra can now handle real terminal tasks instead of just pretending it did them. It even auto-continues unfinished tasks!
• No more “AI Confidence Theater”: Web research is much smarter. Instead of grabbing random snippets, Hydra can now search, inspect pages, and evaluate if it actually found the answer before moving on.

🎙️ Self-Managing Wake Words

Thanks to Firmware v3.0.3, Tater can now change its own satellite wake words!

You can literally say, “Tater, change your wake word to hey computer,”* and it works.

• The new `microwakewordVerba` handles natural spoken phrases better (e.g., “hey tatty” vs “hey-tatty”) and can reset to defaults if things get weird.

📅 Enhanced Personal Core Dashboard

Your dashboard is getting a much-needed productivity boost:

• 7-day outlook context: Get a better grasp of your upcoming schedule.
• Multi-user support: Easily switch between different person profiles.
• Deep Context: Added calendar, action, and subscription context directly into your briefs.

This release marks a massive shift toward Tater being a persistent workflow engine. Happy scripting (and please, use common sense with that shell access)! 🥔⚙️

🔗 View Release

Deep-Live-Cam – 2.7-RC1

🚀 Deep-Live-Cam 2.7-RC1 Update Alert! 🚀

If you’ve been looking for ways to level up your real-time face swapping and live streaming game, this release is a massive leap forward in speed and hardware integration. Whether you’re tinkering with local GPUs or cloud-based diffusion, there is some serious magic happening here.

New Live Modes & Engines:

• Decart Live: You can now stream your webcam directly through Decart’s cloud GPUs using real-time diffusion. Just plug in your API key and choose between the Lucy 2.0 or 2.1 variants! ☁️

FLUX Live (prompt-edit): A powerhouse feature for high-end rigs! This enables real-time, prompt-driven face editing at a smooth 30fps using FluxRT. Warning: You’ll need an RTX 5090 or RTX 6000-class GPU to handle this beast.*

• ReSwapper: A brand new face-swap engine has joined the lineup alongside Inswapper and Hyperswap, giving you more options for your swaps.

UI/UX & Hardware Perks:

• RTX Upscaler: If you have compatible hardware, a new toggle in settings lets you use RTX power to upscale your output beyond the source resolution. 💎
• View Toggles: You can now switch between “Classic” and “Modern” UI layouts with a single click from the status bar.
• Cleaner Live View: The floating player overlay now hides automatically during live streams, giving you a much cleaner viewport for focusing on your content.

Global Accessibility & Polish:

• Full Localization: The interface is officially multilingual! You can switch between English, Chinese, Japanese, French, and Arabic in the settings (just remember to restart the app to apply changes). 🌍
• Improved Window Projection: Resizing during live sessions is now automatic, and a new dedicated close button has been added to the overlay for easier management.

All your favorite tools like RIFE Interpolation, Lip Mask, and Face Enhancer are still baked in and ready to go! 🛠️

🔗 View Release