Models & Labs
Llama.cpp b9116 Release Adds MiMo v2.5 Vision
llama.cpp Releaseshigh confidence
Why it matters
- →MiMo v2.5 enhances vision support, crucial for AI applications involving image processing.
- →The update resolves technical issues, improving reliability and performance.
- →Broad platform support increases accessibility for developers across various systems.
The b9116 release of llama.cpp brings MiMo v2.5, focusing on enhancing vision support with features like fused qkv. This update resolves issues such as f16 vision overflow and includes code cleanups for improved maintenance. The release supports a wide range of platforms, including macOS, Linux, and Windows, making it accessible to developers across different systems. This enhancement in vision capabilities positions llama.cpp as a more versatile tool for AI developers.
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 b9085 Release Enhances MiMo-V2.5
llama.cpp Releases · May 9, 2026 · Same story
What happened next
- 2llama.cpp b9684 Release Adds 3D Convolution
llama.cpp Releases · June 18, 2026 · Same story
More from llama.cpp Releases
Coding Toolscoding
llama.cpp b12022 adds ROCm 10 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally bringing ROCm 10.0 to Linux and Windows alongside CUDA 13.4, effectively closing the hardware gap for AMD users who were previously left behind. The inclusion of native Snapdragon support on Linux marks a significant step toward running local LLMs on ARM-based mobile chips without emulation overhead. While KleidiAI on Apple Silicon is disabled in this build, the broader expansion to diverse accelerators ensures that developers aren't locked into NVIDIA ecosystems anymore.
llama.cpp Releases
Coding Toolscoding
llama.cpp b12023 fixes shared prefix bug
This release addresses a specific inference artifact where decision prefixes were incorrectly bleeding into output embeddings. By capping shared prefixes at the first output position, llama.cpp ensures that option tokens remain distinct for parent and child tasks. This is a targeted fix for multi-step reasoning or voting workflows rather than a performance boost. It stabilizes complex generation patterns without changing the underlying model architecture.
