Change8

b10206

📦 llama-cppView on GitHub →
3 features🔧 1 symbols

Summary

This release enforces consistent K and V cache types for DeepSeek V4 and MLA models. It also enables FA when the V cache is quantized for DeepSeek V4.

✨ New Features

  • Enforce the same K and V cache types for DeepSeek V4.
  • Enable FA if V cache is quantized for DeepSeek V4.
  • Enforce the same K and V cache types for MLA models.

Affected Symbols