github llama.cpp · 1h ago · inference

llama.cpp b10155 adds MiMo-V2.5 audio support

from English

The llama.cpp project has released build b10155, which introduces support for the MiMo-V2.5 audio input model. This update includes a GGUF converter for MiMo audio and implements the necessary C++ backend logic to handle RVQ-based audio models.

Added support for MiMo-V2.5 audio input using an RVQ-based model architecture.
Included a GGUF converter specifically for MiMo audio formats.
Implemented the corresponding C++ inference code for the new model type.
Provided binaries for macOS (Apple Silicon and Intel), Linux, Windows, Android, and openEuler across CPU, GPU, and specialized hardware backends.

Importance 1/3 Trust 1/3 llama.cpp Inference efficiency Local inference

Read original

github llama.cpp · 9h ago

llama.cpp b10150 adjusts ggml offloading logic and fixes dsv4 graph

The llama.cpp project has released build b10150, which includes adjustments to the ggml logic for offloading operations to the weight's backend. This update also addresses fixes for the llama dsv4 graph.

media Hugging Face Forums · 11h ago

TraceML adds Hugging Face Trainer callback to diagnose data-loading bottlenecks

The author has integrated a diagnostic tool into TraceML, an open-source PyTorch observability project, to help users distinguish between GPU compute-bound and input-bound training performance issues. This new feature allows developers to quickly identify whether low GPU utilization is caused by slow data loading or actual model computation limits.

github llama.cpp · 13h ago · 2 views

llama.cpp b10150 adjusts ggml offloading logic and fixes dsv4 graph

TraceML adds Hugging Face Trainer callback to diagnose data-loading bottlenecks

Aaron Teo adds -lm mlock argument to llama.cpp

llama.cpp b10093 fixes DeepSeek v4 template and adds drop_reasoning flag

llama.cpp b10089 adds full k-quant and i-quant support to CUDA GET_ROWS