Skip to content
GPULlama3 Build & Run

[FP16] Improved performance by fusing dequantize with compute in kernels: 20-30% Inference Speedup #144

Sign in to view logs

[FP16] Improved performance by fusing dequantize with compute in kernels: 20-30% Inference Speedup

[FP16] Improved performance by fusing dequantize with compute in kernels: 20-30% Inference Speedup #144

build-and-run (opencl)

build-and-run (opencl)
succeeded Dec 4, 2025 in 6m 58s