diff --git a/batchata/providers/gemini/gemini_provider.py b/batchata/providers/gemini/gemini_provider.py
index 223056a..e7fb34a 100644
--- a/batchata/providers/gemini/gemini_provider.py
+++ b/batchata/providers/gemini/gemini_provider.py
@@ -48,7 +48,7 @@ def __init__(self, auto_register: bool = True):
         if not api_key:
             raise ValueError("GOOGLE_API_KEY environment variable is required")
         
-        self.client = genai_lib.Client(api_key=api_key)
+        self.client = genai_lib.Client(vertexai=False, api_key=api_key)
         super().__init__()
         self.models = GEMINI_MODELS
         self._batches: Dict[str, Dict] = {}
diff --git a/batchata/providers/gemini/models.py b/batchata/providers/gemini/models.py
index 339abd8..dbaf69f 100644
--- a/batchata/providers/gemini/models.py
+++ b/batchata/providers/gemini/models.py
@@ -2,96 +2,67 @@
 
 from ..model_config import ModelConfig
 
+# updated based on current (Jan 28 2026) docs
+# models - https://ai.google.dev/gemini-api/docs/models
+# batch pricing - https://ai.google.dev/gemini-api/docs/pricing
+# image file types - https://ai.google.dev/gemini-api/docs/image-understanding#supported-formats
 
 # Google Gemini models with batch processing support
 # Batch mode provides 50% discount on standard API pricing
 GEMINI_MODELS = {
-    "gemini-3.0-pro-latest": ModelConfig(
-        name="gemini-3.0-pro-latest",
-        max_input_tokens=2097152,  # 2M context
-        max_output_tokens=8192,
-        batch_discount=0.5,  # 50% discount confirmed in docs
-        supports_images=True,
-        supports_files=True,
-        supports_citations=False,
-        supports_structured_output=True,
-        file_types=[".pdf", ".txt", ".jpg", ".png", ".gif", ".webp"]
-    ),
-    "gemini-3.0-pro": ModelConfig(
-        name="gemini-3.0-pro",
-        max_input_tokens=2097152,  # 2M context
-        max_output_tokens=8192,
-        batch_discount=0.5,  # 50% discount confirmed in docs
-        supports_images=True,
-        supports_files=True,
-        supports_citations=False,
-        supports_structured_output=True,
-        file_types=[".pdf", ".txt", ".jpg", ".png", ".gif", ".webp"]
-    ),
-    "gemini-3.0-flash-latest": ModelConfig(
-        name="gemini-3.0-flash-latest",
+    "gemini-3-pro-preview": ModelConfig(
+        name="gemini-3-pro-preview",
         max_input_tokens=1048576,  # 1M context
-        max_output_tokens=8192,
-        batch_discount=0.5,
-        supports_images=True,
-        supports_files=True,
-        supports_citations=False,
-        supports_structured_output=True,
-        file_types=[".pdf", ".txt", ".jpg", ".png", ".gif", ".webp"]
-    ),
-    "gemini-3.0-flash": ModelConfig(
-        name="gemini-3.0-flash",
-        max_input_tokens=1048576,  # 1M context
-        max_output_tokens=8192,
-        batch_discount=0.5,
+        max_output_tokens=65536,
+        batch_discount=0.5,  # 50% discount confirmed in docs
         supports_images=True,
         supports_files=True,
         supports_citations=False,
         supports_structured_output=True,
-        file_types=[".pdf", ".txt", ".jpg", ".png", ".gif", ".webp"]
+        file_types=[".pdf", ".txt", ".jpg", ".png", ".webp"]
     ),
-    "gemini-3.0-flash-lite-latest": ModelConfig(
-        name="gemini-3.0-flash-lite-latest",
+    "gemini-3-flash-preview": ModelConfig(
+        name="gemini-3-flash-preview",
         max_input_tokens=1048576,  # 1M context
-        max_output_tokens=8192,
-        batch_discount=0.5,
+        max_output_tokens=65536,
+        batch_discount=0.5,  # 50% discount confirmed in docs
         supports_images=True,
         supports_files=True,
         supports_citations=False,
         supports_structured_output=True,
-        file_types=[".pdf", ".txt", ".jpg", ".png", ".gif", ".webp"]
+        file_types=[".pdf", ".txt", ".jpg", ".png", ".webp"]
     ),
-    "gemini-3.0-flash-lite": ModelConfig(
-        name="gemini-3.0-flash-lite",
+    "gemini-2.5-pro": ModelConfig(
+        name="gemini-2.5-pro",
         max_input_tokens=1048576,  # 1M context
-        max_output_tokens=8192,
+        max_output_tokens=65536,
         batch_discount=0.5,
         supports_images=True,
         supports_files=True,
         supports_citations=False,
         supports_structured_output=True,
-        file_types=[".pdf", ".txt", ".jpg", ".png", ".gif", ".webp"]
+        file_types=[".pdf", ".txt", ".jpg", ".png", ".webp"]
     ),
-    "gemini-2.0-flash": ModelConfig(
-        name="gemini-2.0-flash",
+    "gemini-2.5-flash": ModelConfig(
+        name="gemini-2.5-flash",
         max_input_tokens=1048576,  # 1M context
-        max_output_tokens=8192,
+        max_output_tokens=65536,
         batch_discount=0.5,
         supports_images=True,
-        supports_files=True,
+        supports_files=False,
         supports_citations=False,
         supports_structured_output=True,
-        file_types=[".pdf", ".txt", ".jpg", ".png", ".gif", ".webp"]
+        file_types=[".txt", ".jpg", ".png", ".webp"]
     ),
-    "gemini-2.0-flash-lite": ModelConfig(
-        name="gemini-2.0-flash-lite",
+    "gemini-2.5-flash-lite": ModelConfig(
+        name="gemini-2.5-flash-lite",
         max_input_tokens=1048576,  # 1M context
-        max_output_tokens=8192,
+        max_output_tokens=65536,
         batch_discount=0.5,
         supports_images=True,
         supports_files=True,
         supports_citations=False,
         supports_structured_output=True,
-        file_types=[".pdf", ".txt", ".jpg", ".png", ".gif", ".webp"]
-    ),
+        file_types=[".pdf", ".txt", ".jpg", ".png", ".webp"]
+    )
 }
\ No newline at end of file