github llama.cpp · just now · inference

llama.cpp b10896 fixes DFlash decoding failure with vision models

from English

The llama.cpp project released build b10896, which addresses a critical issue where the drafter memory failed to allocate new tokens when using DFlash with vision models. This failure occurred because images reported a fixed offset, preventing proper token allocation.

The fix stops copying image offsets to allow the drafter to continue decoding.
M-RoPE skip is limited to images only, allowing audio to pass through unchanged.
Comments were cleaned up to align with the updated implementation logic.

This update ensures that speculative decoding with DFlash functions correctly for multimodal models by resolving memory allocation errors caused by image metadata handling.

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

Read original

github llama.cpp · 2d ago · 2 views

llama.cpp disables lazy tensor loading by default on iGPUs

The llama.cpp project has disabled lazy tensor loading by default for integrated GPUs (iGPUs) in release b10867. This change addresses a regression introduced in the previous "AUTO" lazy mode setting.

github llama.cpp · 2d ago · 4 views

llama.cpp b10864 fixes checkpoint eviction for hybrid models

The llama.cpp project released build b10864, which includes a fix for the checkpoint spacing eviction logic in `create_checkpoint()`. Previously, the eviction rule kept the oldest checkpoint and erased later ones within `checkpoint_min_step`, causing prompts shorter than that threshold to drop the checkpoint needed for resumption. This forced hybrid and recurrent models to unnecessarily re-prefill from the previous checkpoint.

github llama.cpp · 3d ago · 6 views

llama.cpp b10844 adds Vulkan DeepSeek-V4 hyper-connection fused ops

The llama.cpp project released version b10844, introducing support for DeepSeek-V4 hyper-connection fused operations (DSV4_HC_COMB/PRE/POST) in the Vulkan backend. This update brings Vulkan to parity with CUDA and Metal regarding these specific optimizations.

github llama.cpp · 3d ago · 5 views

llama.cpp b10842 adds gfx90c HIP support

The llama.cpp project released build b10842, which introduces support for the gfx90c architecture via HIP.

media Hugging Face Forums · 3d ago · 9 views

AI Model Radar launches with measured hardware fit for open models

A new site called AI Model Radar helps users determine which open models actually fit their GPU or Mac by using measured data rather than estimates. The platform calculates weight sizes based on real published GGUF files and computes context cache layer-by-layer from model configurations.