Gemini 4 Neon wrote AMD kernels that nearly doubled Inkling 2 Mini decode speed in 4.3 hours
The agent-built Loom package also hit 5.9x llama.cpp prefill at an 18k-token prompt, against day-one support the AMD engineer called super nascent.
By Ryan Merket · Published

The agent-built Loom package also hit 5.9x llama.cpp prefill at an 18k-token prompt, against day-one support the AMD engineer called super nascent.