Text Generation Webui – v4.3.2

text-generation-webui v4.3.2 is officially live! 🚀 This Gradio-based powerhouse is the go-to interface for running LLMs locally, and this update brings some serious heavy-hitting performance boosts and expanded model support for all you tinkerers out there.

Here is the breakdown of what’s new in this release:

Core Model & Backend Upgrades

• Gemma 4 Support: You can now run Gemma 4 with full tool-calling capabilities enabled in both the API and the UI. 🆕
• New `ik_llama.cpp` Backend: A massive addition for performance enthusiasts! This backend offers superior KV cache quantization using Hadamard rotation, better optimizations for MoE models, and improved CPU inference.
• Transformers Enhancements: The engine now auto-detects `torch_dtype` from model configs rather than forcing half-precision, making the model loading process much smarter.

API & UI Improvements

• Enhanced Completions API: The `/v1/completions` endpoint now supports `echo` and `logprobs`, allowing you to see token-level probabilities and IDs. 📊
• Snappier Interface: A custom Gradio fork has been optimized to save up to 50ms per UI event, making button clicks and transitions feel much smoother.
• Smarter Templates: Instruction templates are now detected via model metadata instead of relying on old filename patterns.

Security & Stability Fixes

• Hardened Security: Fixed an ACL bypass in the Gradio fork for Windows/macOS and added server-side validation for various input groups like Dropdowns and Radio buttons. 🛡️
• SSRF Protection: Added URL validation to `superbooga` extensions to block requests to private or internal networks.
• Bug Squashing: Resolved several critical issues, including crashes related to Gemma 4 templates in llama.cpp and loading failures for Qwen3.5 MoE models.

Portable Builds & Updates

New self-contained packages are available for Windows, Linux, Mac, and various GPU architectures (NVIDIA CUDA, AMD Vulkan/ROCm, and Intel). If you’re using the portable version, updating is easier than ever—you can now use a shared `user_data` folder across multiple installs! 📂

🔗 View Release

ComfyUI – v0.18.5

ComfyUI v0.18.5 is officially live! 🚀

For those of you building complex, node-based generative AI pipelines, ComfyUI continues to be the powerhouse engine for granular control over Stable Diffusion and beyond. Whether you’re orchestrating intricate image upscaling or multi-step video generation, this tool remains the gold standard for modularity and efficiency.

This latest minor version update focuses on keeping your creative workflows smooth and reliable. Here is what’s new in v0.18.5:

• Enhanced Stability: This patch includes refinements to existing code, specifically aimed at ensuring smoother operation when executing heavy or complex node sequences.
• Core Maintenance: As part of the ongoing development by Comfy-Org, this release ensures the core engine stays perfectly aligned with the rapidly evolving broader AI ecosystem.

If you’ve been pushing your hardware through massive workflows and want to ensure peak performance and stability, now is a great time to pull this update! 🛠️

🔗 View Release

Text Generation Webui – v4.3.1

text-generation-webui v4.3.1 is officially live, and it’s a massive one for anyone looking to push the boundaries of local LLM inference! 🚀

This Gradio-based web UI is essentially the “AUTOMATIC1111” equivalent for text generation, providing a comprehensive interface to run Large Language Models locally with support for multiple backends like llama.cpp, Transformers, and ExLlama.

Here’s what’s new in this release:

• Model & Inference Upgrades:
• 🆕 Gemma 4 Support: Full integration including tool-calling capabilities in both the API and UI.
• ik_llama.cpp Backend: New support via portable builds (or the `–ik` flag for full installs) offering specialized optimizations for MoE models, improved CPU inference, and highly accurate KV cache quantization.
• Transformers Optimization: The UI now auto-detects `torch_dtype` from model configs instead of forcing bf16/f16.
• ExLlamaV3 Fixes: Resolved issues with Qwen3.5 MoE loading and fixed `ban_eos_token` functionality.

• API Enhancements:
• The `/v1/completions` endpoint now supports `echo` and `logprobs` parameters, returning token-level probabilities and new `top_logprobs_ids`.

• Performance & UI Tweaks:
• Snappier Interface: A custom Gradio fork has been optimized to save up to 50ms per UI event (like button clicks).
• Smarter Templates: Instruction templates now detect from model metadata rather than relying on filename patterns.

• Security & Stability:
• Fixed a critical ACL bypass in the Gradio fork for Windows/macOS.
• Added server-side validation for input components (Dropdown, Radio, etc.).
• Patched an SSRF vulnerability in superbooga extensions by validating fetched URLs against private networks.

🛠️ Pro-tip for updating: If you’re using a portable install, just download the latest version and replace your `user_data` folder. Since version 4.0, you can actually keep `user_data` one level up (next to your install folder) to make future updates even smoother!

🔗 View Release

Text Generation Webui – v4.3

🚨 Text-Generation-WebUI v4.3 is live! 🚨

Hey AI tinkerers & devs — fresh update dropped, and it’s packed with performance wins, new backends, and security upgrades. Here’s the lowdown:

—

🔹 🔥 Brand-new backend: `ik_llama.cpp`

A high-octane fork by the imatrix creator, now baked into TGWU:

• ✅ New quant formats (Q4_K_M, Q6_K, etc.)
• 🧠 Hadamard-based KV cache quantization — way more accurate, on by default
• ⚡ Built for MoE models & CPU inference (yes, really fast)

→ Grab it via `textgen-portable-ik` or `–ik` flag!

—

🔹 🧠 API upgrades (OpenAI-compatible!)

The `/v1/completions` endpoint now supports:

• `echo`: Returns prompt + completion in one go
• `logprobs`: Token-level log probabilities (prompt & generated)
• `top_logprobs_ids`: Top token IDs per position — perfect for probing model confidence 🎯

—

🔹 🎨 Gradio UX + Security Boost

• 🐌 Custom Gradio fork = ~50ms faster UI interactions
• 🔒 Fixed ACL bypass (Windows/macOS path quirks)
• ✅ Server-side validation for Dropdown/Radio/CheckboxGroup
• 🛡️ SSRF fix in superbooga: blocks internal/private IPs

—

🔹 🔧 Bug fixes & polish

• `–idle-timeout` now works for encode/decode + parallel generations ✅
• Stopping strings fixed (e.g., `<|return|>` vs `<|result|>`)
• Qwen3.5 MoE loads cleanly via ExLlamaV3_HF
• `ban_eos_token` finally works (EOS suppression at logit level)

—

🔹 📦 Dependency upgrades

• 🦙 `llama.cpp` → latest (`a1cfb64`) + Gemma-4 support
• 🔄 `ExLlamaV3` → v0.0.28
• 📦 `transformers` → 5.5
• ✨ Auto-detects `torch_dtype` from model config (override with `–bf16`)
• 🗑️ Removed obsolete `models/config.yaml` — templates pulled from model metadata now

—

🔹 📌 Terminology update

“Truncation length” → now “context length” in logs (more accurate, less confusing!)

—

🔹 📦 Portable builds — GGUF-ready & zero-install

| Platform | Build to Use |

|———-|————–|

| NVIDIA (old driver) | `cuda12.4` |

| NVIDIA (new driver, CUDA >13) | `cuda13.1` |

| AMD/Intel GPU | `vulkan` |

| AMD (ROCm) | `rocm` |

| CPU-only | `cpu` |

| Apple Silicon | `macos-arm64` |

| Intel Mac | `macos-x86_64` |

🔁 Updating? Just swap the folder — keep `user_data/`, and now you can even move it one level up for shared use across versions 🎉

—

Let me know if you want a quick-start walkthrough on `ik_llama.cpp` or portable builds! 🛠️🚀

🔗 View Release

Ollama – v0.20.0: tokenizer: add byte fallback for SentencePiece BPE encoding (#15232)

🚨 Ollama v0.20.0 is live! 🚨

Big tokenizer upgrade incoming—this one’s a must-have for accuracy and reliability, especially when dealing with non-ASCII or rare characters. Here’s what’s new:

🔹 Byte fallback for SentencePiece BPE

→ When a character can’t be tokenized via standard BPE merges, Ollama now falls back to encoding each UTF-8 byte as a `<0xHH>` token (e.g., `€` → `<0xE2><0x82><0xAC>`).

→ No more silent character loss! 🛡️

🔹 Decoding updated too

→ The decoder now correctly reconstructs the original bytes from `<0xHH>` tokens—ensuring perfect round-trip fidelity. ✅

🔧 Fixes:

• #15229 (dropped chars on encode)
• #15231 (decoder crashes with unknown tokens)

This upgrade boosts robustness across multilingual, technical, or edge-case text—making local LLM inference even more reliable. 🧠⚡

Upgrade now and keep your tokens tight! 📦✨

🔗 View Release

Home Assistant Voice Pe – 26.4.0

🚨 Home Assistant Voice PE 26.4.0 is live! 🚨

Hey AI tinkerers & smart home builders—big update dropped for Home Assistant Voice PE! This open-source voice control gem just got even smoother, more robust, and community-powered 🙌

🔥 What’s New in v26.4.0?

✅ Media Playback Stability Boost 🎧

No more audio dropouts—TTS and media playback now run buttery-smooth, even under load.

✅ Multi-Sendspin Server Support 🌐

Deploy multiple Sendspin instances for redundancy, load balancing, or global reach. Scaling just got way easier!

✅ TTS Timeout Bug Squashed ⏱️

Finally—full TTS responses every time. No more mid-sentence cutoffs. 🎉

✅ New Contributor Welcome! 👏

Shoutout to @akloeckner for their first PR (#558)—the power of open source in action!

💡 Bonus: The project is now officially sponsored by the Open Home Foundation—a huge win for privacy-first, offline-capable voice control!

🔗 Dig into all the nitty-gritty: [Changelog (25.12.4 → 26.4.0)](link-to-full-changelog)

Let’s make voice control truly accessible, local, and open—drop a 🎤 if you’re upgrading!

🔗 View Release

Ollama – v0.20.0-rc1: convert: support new Gemma4 audio_tower tensor naming (#15221)

🚨 Ollama v0.20.0-rc1 is live! 🚨

Hey AI tinkerers — big news for multimodal & audio model fans!

🔹 New in this RC: The `convert` tool now supports the updated tensor naming convention for Gemma 4’s `audio_tower` (PR #15221).

🎯 Why this rocks:

• Fixes tensor-mismatch errors when converting or fine-tuning Gemma 4 audio models (think speech-to-text, multimodal ASR, etc.).
• Keeps Ollama in lockstep with the latest Hugging Face model formats — no more manual tensor renaming hacks!
• Makes it way smoother to run Gemma 4’s audio capabilities locally 🎙️➡️🧠

📦 Note: This is a release candidate — stable enough to test, but expect polish & tweaks before final v0.20.0 drops.

Try it out and let us know how your audio conversions go! 🧪🎧

#Ollama #Gemma4 #AudioAI #LLMLocal

🔗 View Release

Ollama – v0.20.0-rc0: Merge pull request #42 from ollama/jmorganca/gemma4-ggml-improvements

🚨 Ollama v0.20.0-rc0 is here!

PR #42 just landed — bringing sweet, sweet Gemma 4 GGML love 🍬

🔹 Gemma 4 GGML Fixes

• Fixed a critical bug in the MoE fused gate/up projection split, boosting correctness & speed for CPU-based Gemma 4 runs.
• Improved multiline tool-call argument parsing — now handles complex JSON args (e.g., multi-line function/tool outputs) way more reliably.

🎯 Focus: Gemma 4 + GGML optimization & rock-solid tool-use support.

⚠️ Reminder: This is a release candidate (rc0) — great for testing, not quite prod-ready yet. Stay tuned for the final `v0.20.0`! 🚀

🔗 View Release

Wyoming Openai – TTS voice name disambiguation (0.4.3)

🚨 Wyoming OpenAI v0.4.3 is live—and it’s a quiet hero with big improvements! 🚨

🔥 TTS Voice Name Disambiguation (#55)

No more confusing duplicate voice names! If multiple TTS models (e.g., Kokoro, Piper, Edge) share a voice like `”en-us”` or `”alice”`, they now show up uniquely—like `Alice [Kokoro]`, `Alice [Edge]`—so Wyoming clients (like Home Assistant) can pick the exact one you want. 🎯

🛠️ Behind-the-scenes polish

• ✅ `pyright` added for stricter type safety (fewer sneaky bugs!)
• 🔄 CI/CD updated to latest GitHub Actions
• 🐳 Docker builds are leaner with `.dockerignore` and optimized layers
• 📦 All deps bumped—clean, modern, secure

📦 Still installable via `pip install wyoming-openai`, and Docker Compose setups for LocalAI, Edge TTS, Kokoro, Chatterbox & more.

No API keys needed if you’re using local or open-source models—just drop in your preferred TTS/STT backend and go. 🧠➡️🔊

👉 Full details: [v0.4.2…v0.4.3](link-to-changelog)

Happy voice-hacking, folks! 🎙️✨

🔗 View Release

Tater – Tater v69

🚨 Tater v69 — “Eyes Everywhere” 📡👁️🥔 is live, and it’s a visual upgrade that’ll make your local AI see the world like never before!

🔥 Awareness Core: Multi-Provider Vision

✅ UniFi Protect is now a first-class event provider — join Home Assistant in feeding Tater real-time camera/motion data. No hacks, no glue code — just plug-and-play awareness from either system (or both!).

🔄 Hybrid Awareness Mode

👁️ Use UniFi Protect for sensing (cameras, doorbells, motion events)

🧠 Keep Home Assistant for acting (TTS, notifications, media playback)

All stitched together via one unified Awareness layer — smooth, reliable, and yours.

🧠 AI Task Core: Smarter & More Reliable

✔️ Refined task routing & destination handling (think: “remind me in the kitchen” → actually works)

✔️ Fewer hiccups, more dependable automation flow

🎵 ComfyUI Audio ACE: Fixed & Tuned

🎶 Stability fixes + cleaner generation reliability → fewer dropped beats, more vibes 🎧

⚙️ Under-the-Hood Polish

• Cleaner provider abstraction & event pipeline

• General reliability boosts across all cores

🥔 Bottom line:

> Tater v68 gave it eyes. Tater v69 lets it choose which ones to trust.

Same brilliant Hydra brain 🧠 — now with flexible perception 👁️👁️.

Ready to upgrade? 🚀

👉 Check the Tater Shop, update your Core, and let your AI see more. 🥔✨

🔗 View Release