From 3248a3599275c3d2a98bc41d512e1d3ba66e32ba Mon Sep 17 00:00:00 2001 From: Kawrakow Date: Wed, 2 Jul 2025 09:27:47 +0200 Subject: Adding IQ3_KS quants (#566) * iq3_ks: basics * iq3_ks: CUDA dequantize * iq3_ks: CUDA mmvq * iq3_ks: mmq * iq3_ks: faster mmq * iq3_ks: Zen4 * iq3_ks: AVX2 convert to q8_k_r8 This gives usPP-512 = 360 t/s. * iq3_ks: AVX2 GEMM/GEMV * iq3_ks: NEON GEMM/GEMV * iq3_ks: NEON convert to q8_k_r8 This gives us PP-512 = 164 t/s. * iq3_ks: Metal dequantize * iq3_ks: Metal gemv - pathetic performance --------- Co-authored-by: Iwan Kawrakow --- examples/quantize/quantize.cpp | 1 + 1 file changed, 1 insertion(+) (limited to 'examples/quantize/quantize.cpp') diff --git a/examples/quantize/quantize.cpp b/examples/quantize/quantize.cpp index 85ceabfd..fd657373 100644 --- a/examples/quantize/quantize.cpp +++ b/examples/quantize/quantize.cpp @@ -76,6 +76,7 @@ static const std::vector QUANT_OPTIONS = { { "IQ2_K_R4", LLAMA_FTYPE_MOSTLY_IQ2_K_R4, "IQ2_K repacked",}, { "IQ2_KS", LLAMA_FTYPE_MOSTLY_IQ2_KS, " 2.1875 bpw non-linear quantization",}, { "IQ2_KT", LLAMA_FTYPE_MOSTLY_IQ2_KT, " 2.125 bpw trellis quantization", }, + { "IQ3_KS", LLAMA_FTYPE_MOSTLY_IQ3_KS, " 3.19 bpw non-linear quantization", }, { "IQ3_K", LLAMA_FTYPE_MOSTLY_IQ3_K, " 3.44 bpw non-linear quantization", }, { "IQ3_K_R4", LLAMA_FTYPE_MOSTLY_IQ3_K_R4, "IQ3_K repacked", }, { "IQ3_KL", LLAMA_FTYPE_MOSTLY_IQ3_KL, " 4 bpw non-linear quantization mix",}, -- cgit v1.2.3