Skip to content

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.

feat(api-nodes): add Gemini 3.1 Flash Lite model to LLM node (#12803)

Alexander Piskun34e55f006156
diff --git a/comfy_api_nodes/nodes_gemini.py b/comfy_api_nodes/nodes_gemini.py
index d83d2fc15..8225ea67e 100644
--- a/comfy_api_nodes/nodes_gemini.py
+++ b/comfy_api_nodes/nodes_gemini.py
@@ -72,18 +72,6 @@ GEMINI_IMAGE_2_PRICE_BADGE = IO.PriceBadge(
)
-class GeminiModel(str, Enum):
- """
- Gemini Model Names allowed by comfy-api
- """
-
- gemini_2_5_pro_preview_05_06 = "gemini-2.5-pro-preview-05-06"
- gemini_2_5_flash_preview_04_17 = "gemini-2.5-flash-preview-04-17"
- gemini_2_5_pro = "gemini-2.5-pro"
- gemini_2_5_flash = "gemini-2.5-flash"
- gemini_3_0_pro = "gemini-3-pro-preview"
-
-
class GeminiImageModel(str, Enum):
"""
Gemini Image Model Names allowed by comfy-api
@@ -237,10 +225,14 @@ def calculate_tokens_price(response: GeminiGenerateContentResponse) -> float | N
input_tokens_price = 0.30
output_text_tokens_price = 2.50
output_image_tokens_price = 30.0
- elif response.modelVersion == "gemini-3-pro-preview":
+ elif response.modelVersion in ("gemini-3-pro-preview", "gemini-3.1-pro-preview"):
input_tokens_price = 2
output_text_tokens_price = 12.0
output_image_tokens_price = 0.0
+ elif response.modelVersion == "gemini-3.1-flash-lite-preview":
+ input_tokens_price = 0.25
+ output_text_tokens_price = 1.50
+ output_image_tokens_price = 0.0
elif response.modelVersion == "gemini-3-pro-image-preview":
input_tokens_price = 2
output_text_tokens_price = 12.0
@@ -292,8 +284,16 @@ class GeminiNode(IO.ComfyNode):
),
IO.Combo.Input(
"model",
- options=GeminiModel,
- default=GeminiModel.gemini_2_5_pro,
+ options=[
+ "gemini-2.5-pro-preview-05-06",
+ "gemini-2.5-flash-preview-04-17",
+ "gemini-2.5-pro",
+ "gemini-2.5-flash",
+ "gemini-3-pro-preview",
+ "gemini-3-1-pro",
+ "gemini-3-1-flash-lite",
+ ],
+ default="gemini-3-1-pro",
tooltip="The Gemini model to use for generating responses.",
),
IO.Int.Input(
@@ -363,11 +363,16 @@ class GeminiNode(IO.ComfyNode):
"usd": [0.00125, 0.01],
"format": { "approximate": true, "separator": "-", "suffix": " per 1K tokens" }
}
- : $contains($m, "gemini-3-pro-preview") ? {
+ : ($contains($m, "gemini-3-pro-preview") or $contains($m, "gemini-3-1-pro")) ? {
"type": "list_usd",
"usd": [0.002, 0.012],
"format": { "approximate": true, "separator": "-", "suffix": " per 1K tokens" }
}
+ : $contains($m, "gemini-3-1-flash-lite") ? {
+ "type": "list_usd",
+ "usd": [0.00025, 0.0015],
+ "format": { "approximate": true, "separator": "-", "suffix": " per 1K tokens" }
+ }
: {"type":"text", "text":"Token-based"}
)
""",
@@ -436,12 +441,14 @@ class GeminiNode(IO.ComfyNode):
files: list[GeminiPart] | None = None,
system_prompt: str = "",
) -> IO.NodeOutput:
- validate_string(prompt, strip_whitespace=False)
+ if model == "gemini-3-pro-preview":
+ model = "gemini-3.1-pro-preview" # model "gemini-3-pro-preview" will be soon deprecated by Google
+ elif model == "gemini-3-1-pro":
+ model = "gemini-3.1-pro-preview"
+ elif model == "gemini-3-1-flash-lite":
+ model = "gemini-3.1-flash-lite-preview"
- # Create parts list with text prompt as the first part
parts: list[GeminiPart] = [GeminiPart(text=prompt)]
-
- # Add other modal parts
if images is not None:
parts.extend(await create_image_parts(cls, images))
if audio is not None: