mirrors/ollama-for-amd
1
0
Fork 0
mirror of https://github.com/likelovewant/ollama-for-amd.git synced 2025-12-22 06:43:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

add NumGQA

Browse Source
This commit is contained in:
Michael Yang
2023-07-27 14:04:30 -07:00
parent 18ffeeec45
commit ad3a7d0e2c
2 changed files with 3 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
llama/llama.go
View File

@@ -127,6 +127,7 @@ func New(model string, opts api.Options) (*LLM, error) {
params.seed = C.uint(llm.Seed)
params.n_ctx = C.int(llm.NumCtx)
params.n_batch = C.int(llm.NumBatch)
params.n_gqa = C.int(llm.NumGQA)
params.n_gpu_layers = C.int(llm.NumGPU)
params.main_gpu = C.int(llm.MainGPU)
params.low_vram = C.bool(llm.LowVRAM)