tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-08-17 13:40:55 -04:00
Code Issues Packages Projects Releases Wiki Activity

cuda : improve text-generation and batched decoding performance (#3776)

Browse Source 
* cuda : prints wip

* cuda : new cublas gemm branch for multi-batch quantized src0

* cuda : add F32 sgemm branch

* cuda : fine-tune >= VOLTA params + use MMQ only for small batches

* cuda : remove duplicated cuBLAS GEMM code

* cuda : add CUDA_USE_TENSOR_CORES and GGML_CUDA_FORCE_MMQ macros

* build : add compile option to force use of MMQ kernels
This commit is contained in:
Georgi Gerganov
2023-10-27 17:01:23 +03:00
committed by GitHub
parent 34b2a5e1ee
commit 2f9ec7e271
5 changed files with 125 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
llama.h
View File

@@ -178,7 +178,7 @@ extern "C" {
float rope_freq_scale; // RoPE frequency scaling factor, 0 = from model
// Keep the booleans together to avoid misalignment during copy-by-value.
bool mul_mat_q; // if true, use experimental mul_mat_q kernels
bool mul_mat_q; // if true, use experimental mul_mat_q kernels (DEPRECATED - always true)
bool f16_kv; // use fp16 for KV cache, fp32 otherwise
bool logits_all; // the llama_eval() call computes all logits, not just the last one
bool embedding; // embedding mode only