Coding Tools
llama.cpp fixes OpenVINO GPU regressions and CI
llama.cpp Releaseshigh confidence
Why it matters
- →Fixes stateful execution crashes in Qwen3.5 and Gemma-3 on OpenVINO.
- →Resolves GPU plugin defects causing empty outputs or resource errors.
- →Improves stability for AMD and Intel users running local inference.
llama.cpp has released a patch addressing several regressions in its OpenVINO backend, specifically targeting stateful execution failures on CPU and GPU. The update fixes graph scheduling issues with mixed token/embd branches and corrects quantized type handling that previously caused crashes in models like Qwen3.5 and Gemma-3. Additionally, the release resolves GPU plugin defects related to RMS norm ranks and resource allocation errors during operator tests. These changes ensure more stable inference for users relying on OpenVINO for non-NVIDIA hardware.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1Llama.cpp b10981 release enhances OpenVINO support
llama.cpp Releases · September 16, 2026 · Same story
- 2llama.cpp b11374 optimizes OpenVINO MoE performance
llama.cpp Releases · October 4, 2026 · Same story
Follow this story
Open the full story →llama.cpp b11381 release with CUDA 13 and ROCm 10
2 developments
- Oct 4 · llama.cpp Releases
llama.cpp b11381 release with CUDA 13 and ROCm 10 - Oct 6 · llama.cpp Releases
llama.cpp fixes OpenVINO GPU regressions and CI (This article)↳ llama.cpp patches critical stability issues in its OpenVINO backend, fixing stateful inference crashes on CPU and GPU.
More from llama.cpp Releases
Coding Toolscoding
llama.cpp b11424 adds CUDA 13 and ROCm 10.0
This release quietly closes the hardware gap for local inference by adding default support for CUDA 13 and ROCm 10.0 alongside existing CUDA 12 builds. NVIDIA users can now leverage newer driver stacks without manual configuration, while AMD GPU owners finally get first-class parity with the same ease of use previously reserved for CUDA. Apple Silicon KleidiAI is disabled in this specific build, a notable regression for Mac users who rely on that optimization. The inclusion of Snapdragon and OpenVINO binaries further broadens the reach to edge devices and Intel hardware. It’s less about new features and more about llama.cpp solidifying its position as the universal runtime for every major accelerator.
llama.cpp Releases
Coding Toolscoding
llama.cpp b11425 adds ROCm 10 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally bringing first-class ROCm 10.0 support to both Linux and Windows. AMD GPU users no longer need workarounds, effectively closing a long-standing parity gap with NVIDIA's CUDA ecosystem. The inclusion of Snapdragon AI stack binaries for Linux marks a strategic push into ARM-based edge devices, while the simultaneous addition of CUDA 13 builds ensures compatibility with the latest driver stacks. By standardizing these hardware backends across major operating systems, the project removes friction for developers deploying models on diverse non-NVIDIA hardware.
llama.cpp Releases
Coding Toolscoding
llama.cpp 0.6.0 adds CUDA 13 and Snapdragon support
The llama.cpp 0.6.0 release quietly expands hardware coverage where it counts most: next-gen NVIDIA GPUs and mobile silicon. By shipping native builds for CUDA 13.4 alongside the existing CUDA 12 binaries, users can finally leverage newer GPU architectures without compiling from source. The inclusion of Linux arm64 support for Snapdragon chips with Adreno GPU and Hexagon NPU acceleration signals a serious push into on-device inference beyond Apple Silicon. While KleidiAI on macOS is temporarily disabled, the broader platform expansion makes this one of the most versatile local inference releases in recent memory.
llama.cpp Releases
More in Coding Tools
Coding Toolscoding
Claude Code v2.1.290: Plugin hooks and agent stability fixes
This release significantly tightens the security model for Claude Code plugins by exposing server tool IDs and approval ceilings to hook functions, allowing developers to build more granular permission checks. It also stabilizes long-running agent sessions by fixing critical bugs in subagent resume logic and scheduled task persistence after compaction. For plugin authors, the new validation flags ensure gating hooks are properly configured before deployment. These changes make the platform safer for enterprise use while reducing friction for complex automated workflows.
Claude Code Releases
© GitHub ChangelogCoding Toolscoding
GitHub Secret Scanning adds Lovable and Supabase detectors
GitHub quietly expanded its secret scanning partnership program to include Lovable Labs, Pydantic Services, and Supabase. This update means credentials from these popular development platforms are now automatically detected in public repositories, allowing the providers to revoke or rotate compromised keys before abuse occurs. For developers using Supabase or Lovable, this adds a critical layer of automated security hygiene without requiring manual configuration. It reflects GitHub's ongoing effort to integrate directly with the modern AI and database tooling stack that dominates current development workflows.
GitHub Changelog
© The Verge AICoding Toolsother
Open-source tool deletes Apple Intelligence data on macOS
Apple removed the simple toggle to disable AI features in macOS 27, leaving 12GB of models lingering on disk even when turned off. RemoveMacAI solves this by automating the cleanup: it disables Siri, Writing Tools, and Genmoji, deletes the underlying models, and blocks future downloads. This restores user control over storage and privacy without requiring manual navigation through scattered settings panes. It’s a practical patch for an ecosystem change that prioritized feature retention over user choice.
The Verge AI