Diagnostics

Every risky code path has a URL switch that falls back to the slower, simpler one. Handy to check whether an optimization is responsible for something odd — the answer should be identical, only slower.

?gemv=0        decode matmul → row kernels
?f16shared=0   f16 prefill GEMM → one row per thread
?qshared=0     q4/q8 prefill GEMM → 4 rows per invocation
?qshared2=0    q4/q8 prefill GEMM → v1 tiles (32×64)
?warmup=0      no weight warm-up (first message pays it)
?ggufstream=0  GGUF as one download instead of ranges
?kvq=0         KV cache in f32 instead of int8
?timing=1      per-stage timing of the forward pass, in the console

Add a switch to the /chat URL, reload, and compare. Several can be combined with &.

Brimkern — open WebGPU engine, built by Romain Khanoyan. Local AI, WebGPU, on-device engines.