ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-02-25 07:34:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

iq2_kt: CUDA dequantize

Browse Source 
so we can run perplexity calcs.
As already indicated by rmse, the 2-bit trellis approach is
quite a bit worse than iq2_xxs.
This commit is contained in:
Iwan Kawrakow
2024-11-06 08:34:58 +02:00
parent a4f1ac8da4
commit 426a6e685f
10 changed files with 81 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
examples/quantize/quantize.cpp
View File

@@ -45,6 +45,7 @@ static const std::vector<struct quant_option> QUANT_OPTIONS = {
{ "IQ4_KSS", LLAMA_FTYPE_MOSTLY_IQ4_KSS, " 4.0 bpw non-linear quantization", },
{ "IQ2_K", LLAMA_FTYPE_MOSTLY_IQ2_K, " 2.375 bpw non-linear quantization",},
{ "IQ2_KS", LLAMA_FTYPE_MOSTLY_IQ2_KS, " 2.1875 bpw non-linear quantization",},
{ "IQ2_KT", LLAMA_FTYPE_MOSTLY_IQ2_KT, " 2.0625 bpw non-linear quantization",},
{ "IQ3_K", LLAMA_FTYPE_MOSTLY_IQ3_K, " 3.44 bpw non-linear quantization", },
{ "IQ3_KL", LLAMA_FTYPE_MOSTLY_IQ3_KL, " 4 bpw non-linear quantization mix",},
{ "IQ4_K", LLAMA_FTYPE_MOSTLY_IQ4_K, " 4.5 bpw non-linear quantization", },