matt/ollama37
1
0
Fork 0
mirror of https://github.com/dogkeeper886/ollama37.git synced 2025-12-12 08:47:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

Bump llama.cpp to b1662 and set n_parallel=1

Browse Source
This commit is contained in:
Daniel Hiltgen
2023-12-14 10:25:12 -08:00
parent 89bbaafa64
commit 9adca7f711
3 changed files with 9 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
llm/ext_server.go
View File

@@ -160,7 +160,7 @@ func newExtServer(server extServer, model string, adapters, projectors []string,
sparams.n_batch = C.uint(opts.NumBatch)
sparams.n_gpu_layers = C.int(numGPU)
sparams.main_gpu = C.int(opts.MainGPU)
sparams.n_parallel = 2 // TODO - wire up concurrency
sparams.n_parallel = 1 // TODO - wire up concurrency
// Always use the value encoded in the model
sparams.rope_freq_base = 0.0