MLX-LM – v0.30.0

MLX LM v0.30.0 is live 🚀 — Apple Silicon LLMs just got a serious power-up!

• Server performance fixed: No more busy-waiting — idle polling is now lean, quiet, and efficient. 🛠️
• Transformers v5 fully supported: All the latest tokenizer tweaks, model updates, and Hugging Face magic? Covered. 🤖
• MIMO v2 Flash enabled: Multi-input models now fly with optimized attention — faster inference, less latency. ⚡
• Better error messages: Batching failed? Now you’ll know why — no more cryptic crashes. 📢
• Model parallel generation: Split massive models across GPUs like a pro. Scale your LLMs without rewriting code. 🧩
• Chat template fixes: `apply_chat_template` finally wraps correctly — no more dict chaos in your prompts. ✨

Thousands of Hugging Face models, quantized, fine-tuned, and served — all on your M-series chip. Time to upgrade and push your AI stack further. 🚀

🔗 View Release

Ollama – v0.13.5

🚀 Ollama v0.13.5 just dropped — and it’s a quiet game-changer for Gemma users!

Now you can use function calling with Gemma 2B locally — yes, really. Trigger webhooks, query databases, fetch weather, or call APIs directly from your tiny-but-mighty local Gemma model. No cloud needed.

💡 Why it’s cool:

• Function calling was already in Llama 3 & Mistral — now Gemma joins the party.
• Perfect for building private, lightweight AI agents that do stuff, not just chat.

Under the hood: parser fixes + smoother rendering = fewer hiccups, more flow.

Upgrade in one line:

“`bash

ollama pull gemma:2b

“`

Go build something that acts — not just responds. 🎯

🔗 View Release

Ollama – v0.13.5-rc1

🚀 Ollama v0.13.5-rc1 just dropped — and it’s a game-changer for Gemma users!

Now you can use function calling with Gemma 2B 🎯

Gemma can now dynamically invoke external tools and APIs — think real-time data lookup, code execution, or API calls — all triggered by natural language. No more clunky workarounds. Just define your tools, and let Gemma call them like a pro agent.

Plus:

• Smoother JSON parsing for tool definitions (no more malformed calls!)
• Minor performance boosts and bug fixes
• RC1 = stable enough for early adopters to test in real workflows

Grab it: `ollama pull gemma:2b` and start building smarter, reactive agents today.

Full release notes coming soon — but this feature? Totally worth the upgrade. 🛠️

🔗 View Release

ComfyUI – v0.5.1

ComfyUI v0.5.1 just dropped — and it’s a quiet hero 🛠️

Fixed critical crashes in custom nodes & image loading (no more mid-generate soul-crushes).

Memory management got a serious upgrade — smoother runs on weaker GPUs, even with massive workflows.

Error messages now actually tell you what went wrong… no more “something broke” mysteries.

UI feels slicker: cleaner node links, buttery drag-and-drop.

And hey — partial WebGPU support for Apple Silicon users is now live 🍎⚡ (Big things coming in v0.6!)

If you’ve been battling crashes or lag, this is your sign to update. Keep building! 🚀

🔗 View Release

Ollama – v0.13.5-rc0: GGML update to ec98e2002 (#13451)

Ollama v0.13.5-rc0 just dropped — and it’s all about speed under the hood! 🚀

The GGML inference engine got a major upgrade to commit `ec98e2002`, with smarter, leaner internals:

• ✅ MaskBatchPadding removed — Less padding = less overhead. KQ masking is now cleaner and faster.
• 🚫 NVIDIA Nemotron 3 Nano support paused — Temporarily pulled for stability. Coming back stronger soon!
• 🔧 Solar Pro tweaks — Under-the-hood adjustments, still being verified. If you’re using Solar, test your models!

No flashy UI — just a lighter, faster engine for local LLM inference. Think of it like swapping your car’s engine for a turbocharged version that runs cooler.

Pro tip: Custom models? Run sanity checks — GGML changes can ripple through quantization and attention layers.

Stay sharp, tinkerers. The local LLM revolution keeps accelerating. 🛠️

🔗 View Release

Lemonade – v9.1.1

🔥 Lemonade v9.1.1 just dropped — and it’s a game-changer for local LLM folks!

• New AI features live: Embeddings, reranking, and transcriptions are now built right in — smarter search, better relevance, and full speech-to-text without leaving the app. 🎙️🧠
• ROCm support for Strix Point (gfx1150): AMD Ryzen AI users — your hardware is finally fully optimized. No more waiting, just raw performance.
• Install & config smoothed out: FLM install bugs fixed, `–extra-models-dir` added for custom models, and CMake warnings? Gone.
• UI/UX glow-up: Not-found errors are now helpful, site icons load reliably, and the logo wall got a slick refresh.
• Bug slaying squad in action: Zero-byte downloads? Squashed. Model registration glitches? Fixed. Health request spam? Suppressed.

Plus, the README and install guides were completely rewritten — clearer than ever for newcomers.

Smarter. Faster. More open. 🚀

Time to run your LLMs like a pro — locally, and at full speed.

🔗 View Release

Home Assistant Voice Pe – 25.12.0

Home Assistant Voice PE just dropped v25.12.0 🚀 — and it’s a game-changer for offline voice control!

Now backed by the Open Home Foundation, this ESPHome-powered gem lets you command your smart home without the cloud — perfect for privacy-first tinkerers.

Biggest update? Music Assistant’s Sendspin multi-room audio support (public preview)! Sync tunes across speakers in real-time — think whole-home playlists, party mode activated by voice. 🎶

Shoutout to @theHacker for their first contribution — welcome to the crew! 🙌

78 releases strong and growing.

Full changelog: 25.11.0…25.12.0

Your home just got smarter… and louder.

🔗 View Release

ComfyUI – v0.5.0

ComfyUI v0.5.0 just landed — and it’s a game-changer for workflow builders 🎨✨

• LatentUpscale Node — Upscale in latent space first for smoother, cleaner details with less noise.
• Smarter Node Search — Type “upscale” and it gets you. No more scrolling through 50 nodes.
• Memory Management Upgrade — Fewer crashes, better GPU predictability when running big models.
• Custom Nodes Overhauled — Install, update, and manage third-party nodes like a pro. No more “why isn’t it showing up?!”
• Dark Mode 2.0 — Sleeker, higher contrast. Your eyes won’t scream after midnight prompt sessions.
• Save as Template — Turn your favorite chains into reusable templates. Perfect for teams or recurring styles.

Plus: 40+ bugs squashed, including the dreaded “random node disconnects on reload.” 🐞✅

If you’re deep in Stable Diffusion workflows — this update is your new best friend. Go tweak, test, and create! 🚀

https://www.comfy.org/

🔗 View Release

MLX-LM – v0.29.0

🚀 MLX LM v0.29.0 is live — and it’s a beast!

• Batch generation just got 2x faster thanks to `wired_limit` fixes — your server will thank you.
• RoPE & SuScaledRoPE fixed for `rnj-1` and others — smoother attention, less drift.
• Dequantize bug squashed ✅ Now using the right function — cleaner outputs, better precision.
• Repetition penalty defaults to 0.0 — less annoying repetition from day one. 🎯
• DSV32 & Gemma3 — bugs gone, stable and ready to deploy.
• SSM batching fixed — state-space models now behave on the server. 💡
• Nemotron 3 added! 🎉 Go ahead, test it.
• Devstral-2 now works properly — no more surprises. 👏

Big shoutout to first-time contributors: @otarkhan, @devnamrits, @DePasqualeOrg, and @inferencers — welcome to the crew! 🙌

Update now — your LLMs are ready for a speed run. 🛠️

Full changelog: [v0.28.4…v0.29.0](link)

🔗 View Release

Ollama – v0.13.4

🚀 Ollama v0.13.4 just dropped — tiny update, big impact for Nemotron users!

Fix: The `think` token now actually listens to your custom prompts. No more ignored reasoning directives — if you told the model to “think step by step,” it finally will. 🤔✨

Perfect for RAG builders, prompt engineers, and anyone relying on structured reasoning. No new features, just cleaner, more reliable internal reflection.

Upgrade if you’re using Nemotron models — your prompts just got smarter. 🛠️

🔗 View Release