mirror of
https://github.com/ollama/ollama.git
synced 2025-11-12 09:08:11 +01:00
llm: Enable flash attention by default for gemma3
This commit is contained in:
@@ -893,6 +893,7 @@ func (f GGML) SupportsFlashAttention() bool {
|
||||
// FlashAttention checks if the model should enable flash attention
|
||||
func (f GGML) FlashAttention() bool {
|
||||
return slices.Contains([]string{
|
||||
"gemma3",
|
||||
"gptoss", "gpt-oss",
|
||||
"qwen3",
|
||||
"qwen3moe",
|
||||
|
||||
Reference in New Issue
Block a user