← Back to events
ActiveAIRelease0.34.4-rc0

v0.34.4-rc0: mlx: speed up Qwen 3.8 prompt processing (#18550)

What happened

mlx: speed up Qwen 3.8 prompt processing Use MLX's gated-delta kernel for long scans and fold dense MLP global scales into SwiGLU. address comments

Summary assembled by rule from the sources below

Why it's spreading

Sources