llm: Enable flash attention by default for gemma3

c3c85aa0 · Jesse Gross · Jesse Gross · 0d713051 · c3c85aa0
Commit c3c85aa0 authored Oct 15, 2025 by Jesse Gross Committed by Jesse Gross Oct 15, 2025
Show whitespace changes
Inline Side-by-side

Showing with 1 addition and 0 deletions

fs/ggml/ggml.go fs/ggml/ggml.go +1 -0

No files found.
--- a/fs/ggml/ggml.go
+++ b/fs/ggml/ggml.go
@@ -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",