ggml/llama.cpp releases: b10598

TL;DR

The ggml/llama.cpp project updated its codebase to use a pillow-accurate resizing algorithm across all models and implemented speed optimizations, enhancing both accuracy and performance. These changes are significant for users needing precise model scaling and improved computational efficiency.

Detailed Summary

The ggml/llama.cpp project released version b10598, which includes updates to use the pillow-accurate resizing algorithm across all models and introduces speed optimizations. This update was made by mtmd through pull request #27594. The broader impact of these changes is expected to improve model performance and accuracy in image processing tasks, enhancing user experience on platforms like llama.app.

Key Points

• • mtmd: use pillow-accurate resize algo, correct resize_algo for all models
• • speed optimization