ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-01-26 17:20:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

Q4_K_R4 (#129)

Browse Source 
* Something is still wrong

* Simply don't see what is wrong

* q4_k_r4: finally works on Zen4

I had forgotten to prevent token_embd.weight being quantized
with q4_k_r4!

* q4_k_r4: AVX2

We get PP-512(LLaMA-3.1-8B) = 267 t/s on a Ryzen-5975WX.
This is ~30% better than Q4_K_S.

* q4_k_r4: NEON

We get PP-512(LLaMA-3.1-8B) = 110 t/s.
Not quite as good as q4_0_r4, but still a massive
improvement compared to he 69 t/s for q4_K.

* q4_k_r4: slightly better AVX2

PP-512 goes from 267 t/s to 282 t/s on Ryzen-5975WX

* Minor

* Minor

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
This commit is contained in:
Kawrakow
2024-12-09 16:59:18 +01:00
committed by GitHub
parent b39bbb0405
commit 13126ce100
10 changed files with 465 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
examples/quantize/quantize.cpp
View File

@@ -58,6 +58,7 @@ static const std::vector<struct quant_option> QUANT_OPTIONS = {
{ "IQ5_K", LLAMA_FTYPE_MOSTLY_IQ5_K, " 5.5 bpw non-linear quantization", },
{ "IQ6_K", LLAMA_FTYPE_MOSTLY_IQ6_K, " 6.6 bpw non-linear quantization", },
{ "Q4_K", LLAMA_FTYPE_MOSTLY_Q4_K_M, "alias for Q4_K_M", },
{ "Q4_K_R4", LLAMA_FTYPE_MOSTLY_Q4_K_R4, "Q4_K_S repacked", },
{ "Q4_K_S", LLAMA_FTYPE_MOSTLY_Q4_K_S, " 3.59G, +0.0992 ppl @ LLaMA-v1-7B", },
{ "Q4_K_M", LLAMA_FTYPE_MOSTLY_Q4_K_M, " 3.80G, +0.0532 ppl @ LLaMA-v1-7B", },
{ "Q5_K", LLAMA_FTYPE_MOSTLY_Q5_K_M, "alias for Q5_K_M", },