highperfocused/ollama
1
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2025-04-06 19:08:27 +02:00
Code Issues Packages Projects Releases Wiki Activity

fix metal gpu

Browse Source
This commit is contained in:
Michael Yang 2024-04-02 16:06:45 -07:00
parent a57818d93e
commit 80163ebcb5
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
llm/server.go
View File

@ -85,7 +85,7 @@ func NewLlamaServer(model string, adapters, projectors []string, opts *api.Optio
graph := int64(ggml.KV().GQA()) * kv / 6
usedMemory += graph
if usedMemory > availableMemory || slices.Contains(cpuOnlyFamilies, ggml.KV().Architecture()) {
if (usedMemory > availableMemory || slices.Contains(cpuOnlyFamilies, ggml.KV().Architecture())) && info.Library != "metal" {
info.Library = "cpu"
}
@ -159,7 +159,7 @@ func NewLlamaServer(model string, adapters, projectors []string, opts *api.Optio
params = append(params, "--log-disable")
}
if opts.NumGPU > 0 {
if opts.NumGPU >= 0 {
params = append(params, "--n-gpu-layers", fmt.Sprintf("%d", opts.NumGPU))
}