wehub-resource-sync
16 lines
1.1 KiB
TOML
16 lines
1.1 KiB
TOML
name = "Databricks"
|
|
npm = "@ai-sdk/openai-compatible"
|
|
# Raw Chat is POST `/serving-endpoints/{model}/invocations`: GPT/Gemini 3 use
|
|
# `reasoning_effort`; Claude/Gemini 2.5 use
|
|
# `thinking = { type = "enabled", budget_tokens = N }`.
|
|
# Native OpenAI Responses is POST `/serving-endpoints/responses` and uses
|
|
# `reasoning.effort`; Open Responses supports Claude, Gemini, and open models.
|
|
# Native Gemini `generateContent` uses `generationConfig.thinkingConfig`.
|
|
# https://docs.databricks.com/aws/en/machine-learning/model-serving/query-reason-models (accessed 2026-06-25)
|
|
# https://docs.databricks.com/aws/en/machine-learning/foundation-model-apis/api-reference (accessed 2026-06-25)
|
|
# https://docs.databricks.com/aws/en/machine-learning/model-serving/query-open-responses-models (accessed 2026-06-25)
|
|
# https://docs.databricks.com/aws/en/generative-ai/external-models/google-gemini (accessed 2026-06-25)
|
|
api = "https://${DATABRICKS_HOST}/ai-gateway/mlflow/v1"
|
|
env = ["DATABRICKS_HOST", "DATABRICKS_TOKEN"]
|
|
doc = "https://docs.databricks.com/aws/en/machine-learning/foundation-models/"
|