A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 13, 22:00 UTC
[Misc] Provide Siglip2 chat template (#27939)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Cyrus Leung committed
Nov 2, 2025 at 21:42 UTC
6c317a656eb09a641d85be05aa8498ff160bf0c1
1 file changed
+1
vllm/transformers_utils/chat_templates/registry.py
+1
@@ -40,6 +40,7 @@ _MODEL_TYPE_TO_CHAT_TEMPLATE_FALLBACK: dict[str, ChatTemplatePath] = {
40
"paligemma": CHAT_TEMPLATES_DIR / "template_basic.jinja",
41
"qwen": _get_qwen_chat_template_fallback,
42
"siglip": CHAT_TEMPLATES_DIR / "template_basic.jinja",
43
+ "siglip2": CHAT_TEMPLATES_DIR / "template_basic.jinja",
44
}
45
46