diff options
| author | Kawrakow <iwankawrakow@gmail.com> | 2024-12-14 09:24:30 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-12-14 09:24:30 +0100 |
| commit | 20758edcae65213b2f575b6d23dfea67ad9dd0e0 (patch) | |
| tree | f9f32d541da8bb945a45bbf473b9295496ec5c2b /examples/quantize/quantize.cpp | |
| parent | 12f962dd2494b743deb1c671974a591fdef1f003 (diff) | |
Q8_K_R8: Fastest quantized matrix multiplications (#141)
* q8_k_r8: fastest matrix multiplication known to human kind
We get PP-512(LLaMA-3.1-8B) = 370 t/s on a Ryzen-7950X!
* q8_k_r8: AVX2
I was worried that we don't have enough vector registrers on
AVX2, but it looks like it handles it just fine. We get
PP-512(LLaMA-3.1-8B) = 354 t/s on a Ryzen-5975WX.
Slightly slower than the Zen4 version with double the threads,
but still a huge upgrade compared to Q8_0_R4.
* q8_k_r4: NEON
We get PP-512(LLaMA-3.1-8B) = 159.2 t/s.
Compare this to the 128 t/s we have fr Q8_0_R4.
* q8_k_r4: go to signed ints
Why?
* On AVX2 _mm256_maddubs_epi16() may overflow, so we need to
stay within the signed int range and use _mm256_sign_epi8.
Not yet tested on the AVX2 comp, vut expect major slowdown.
* It is almost 10% faster on ARM_NEON. Somehow the veorrq_u8()
needed tto convert from unsigned to signed seems to be extremely
slow on the M2-Max
* We only lose ~0.5% in oerformance on Zen4 (there the exclusive
or that we now use to convert fro signed to unsigned seems to be
much faster than on M2-Max)
* Shutup useless compiler warnings
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Diffstat (limited to 'examples/quantize/quantize.cpp')
| -rw-r--r-- | examples/quantize/quantize.cpp | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/examples/quantize/quantize.cpp b/examples/quantize/quantize.cpp index f1e3902c..0f906b83 100644 --- a/examples/quantize/quantize.cpp +++ b/examples/quantize/quantize.cpp @@ -70,6 +70,7 @@ static const std::vector<struct quant_option> QUANT_OPTIONS = { { "Q5_K_M", LLAMA_FTYPE_MOSTLY_Q5_K_M, " 4.45G, +0.0122 ppl @ LLaMA-v1-7B", }, { "Q6_K", LLAMA_FTYPE_MOSTLY_Q6_K, " 5.15G, +0.0008 ppl @ LLaMA-v1-7B", }, { "Q6_K_R4", LLAMA_FTYPE_MOSTLY_Q6_K_R4, "Q6_K repacked", }, + { "Q8_K_R8", LLAMA_FTYPE_MOSTLY_Q8_K_R8, "Q8_K repacked", }, { "Q8_0", LLAMA_FTYPE_MOSTLY_Q8_0, " 6.70G, +0.0004 ppl @ LLaMA-v1-7B", }, { "Q4_0_4_4", LLAMA_FTYPE_MOSTLY_Q4_0_4_4, " 4.34G, +0.4685 ppl @ Llama-3-8B", }, { "Q4_0_4_8", LLAMA_FTYPE_MOSTLY_Q4_0_4_8, " 4.34G, +0.4685 ppl @ Llama-3-8B", }, |