feat(databricks): update model YAMLs [bot] (#1760)

Co-authored-by: Truefoundry Models Bot <truefoundry-models-bot@users.noreply.github.com>

Author: models-bot[bot]

providers/databricks/databricks-bge-large-en.yaml

@@ -16,6 +16,7 @@ model: databricks-bge-large-en
 provisioning: serverless
 removeParams:
     - max_tokens
+    - reasoning_effort
 status: active
 supportedModes:
     - embedding

providers/databricks/databricks-claude-opus-4-1.yaml

@@ -20,6 +20,8 @@ modalities:
 mode: chat
 model: databricks-claude-opus-4-1
 provisioning: serverless
+removeParams:
+    - reasoning_effort
 status: active
 supportedModes:
     - chat

providers/databricks/databricks-claude-opus-4-5.yaml

@@ -1,11 +1,15 @@
 costs:
-    - input_cost_per_token: 0.000005
+    - cache_creation_input_token_cost: 6.25e-6
+      cache_read_input_token_cost: 5e-7
+      input_cost_per_token: 0.000005
       output_cost_per_token: 0.000025
       region: "*"
 features:
     - function_calling
     - tool_choice
     - assistant_prefill
+    - prompt_caching
+    - system_messages
 limits:
     context_window: 200000
     max_input_tokens: 200000
@@ -20,6 +24,9 @@ modalities:
 mode: chat
 model: databricks-claude-opus-4-5
 provisioning: serverless
+sources:
+    - https://www.databricks.com/product/pricing/proprietary-foundation-model-serving
+    - https://learn.microsoft.com/en-us/azure/databricks/machine-learning/foundation-model-apis/limits
 status: active
 supportedModes:
     - chat

providers/databricks/databricks-gemma-3-12b.yaml

@@ -19,6 +19,8 @@ modalities:
 mode: chat
 model: databricks-gemma-3-12b
 provisioning: serverless
+removeParams:
+    - reasoning_effort
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/limits
     - https://docs.databricks.com/aws/en/machine-learning/model-serving/function-calling

providers/databricks/databricks-gpt-5-mini.yaml

@@ -1,10 +1,17 @@
 costs:
-    - input_cost_per_token: 2.5e-7
+    - cache_creation_input_token_cost: 2.5e-7
+      cache_read_input_token_cost: 2.5e-8
+      input_cost_per_token: 2.5e-7
       output_cost_per_token: 0.000002
       region: "*"
 features:
     - function_calling
     - tool_choice
+    - structured_output
+    - json_output
+    - prompt_caching
+    - system_messages
+    - cache_control
 limits:
     context_window: 400000
     max_input_tokens: 272000
@@ -32,6 +39,8 @@ provisioning: serverless
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/limits
     - https://docs.databricks.com/aws/en/machine-learning/model-serving/function-calling
+    - https://docs.databricks.com/aws/en/machine-learning/model-serving/structured-outputs
+    - https://www.databricks.com/product/pricing/proprietary-foundation-model-serving
 status: active
 supportedModes:
     - chat

providers/databricks/databricks-llama-4-maverick.yaml

@@ -7,8 +7,6 @@ features:
     - tool_choice
     - structured_output
 limits:
-    context_window: 128000
-    max_input_tokens: 128000
     max_output_tokens: 8192
     max_tokens: 8192
 modalities:
@@ -20,6 +18,8 @@ modalities:
 mode: chat
 model: databricks-llama-4-maverick
 provisioning: serverless
+removeParams:
+    - reasoning_effort
 sources:
     - https://docs.databricks.com/aws/en/machine-learning/model-serving/score-foundation-models
     - https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/limits