[FEATURE]: Native Oracle OCI Generative AI provider support

[kineticquant]

Feature hasn't been suggested before.

• I have verified this feature I'm about to request hasn't been suggested before.

Describe the enhancement you want to request

Requesting to add access to OCI models. OCI offers models by region and is OpenAI-compatible, connecting to
inference.generativeai.{region}.oci.oraclecloud.com (ex: https://inference.generativeai.us-chicago-1.oci.oraclecloud.com/openai/v1).

There is a current pain point when trying to use the custom adapter that prevents usage there, resulting in max_tokens vs max_completion_tokens incompatibility with the current @ai-sdk/openai-compatible adapter.
The error: "Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead."

Appears to happen with any model on OCI, so requesting as feature to add OCI as a provider instead.

[github-actions]

This issue might be a duplicate of existing issues. Please check:

• openai-compatible adapter sends max_tokens to GPT-5/o-series reasoning models that require max_completion_tokens #25096: Same max_tokens vs max_completion_tokens incompatibility with @ai-sdk/openai-compatible adapter — describes the exact same error and root cause, just via a different provider endpoint
• Azure GPT-5 model: "max_tokens" parameter not supported, need "max_completion_tokens" instead #22623: Same error with Azure GPT-5 models requiring max_completion_tokens instead of max_tokens

The underlying fix (making the openai-compatible adapter send max_completion_tokens for models that require it) may resolve your OCI issue without needing a dedicated native provider. It may be worth commenting on #25096 with your OCI-specific details to consolidate.