max_completion_tokens
A parameter introduced by OpenAI in reasoning models (the o1 series and later) that constrains the token limit for an entire completion, including both reasoning and output.
Timeline (last 90 days)
Different vendors' model list endpoints differ in parameter naming, for example some use max_tokens and others use max_completion_tokens.
Cline SDK v0.0.95 modified the parameter handling logic for OpenAI-compatible requests: for the model IDs gpt-6-sol, gpt-6.1-sol, gpt-6-1-sol and gpt-5 through gpt-9, the SDK now sends max_completion_tokens instead of max_tokens, in order to stop OpenAI and Azure from rejecting these requests.
In the OpenAI Chat Completions API, max_tokens is an early field used to limit the total number of tokens in the generated content; OpenAI introduced max_completion_tokens in the o1 series and later reasoning models to constrain the token upper limit of the entire completion, and gradually recommends that new models use the latter.
All Facts (3)
Different vendors' model list endpoints differ in parameter naming, for example some use max_tokens and others use max_completion_tokens.
50%UnverifiedCline SDK v0.0.95 modified the parameter handling logic for OpenAI-compatible requests: for the model IDs gpt-6-sol, gpt-6.1-sol, gpt-6-1-sol and gpt-5 through gpt-9, the SDK now sends max_completion_tokens instead of max_tokens, in order to stop OpenAI and Azure from rejecting these requests.
50%UnverifiedIn the OpenAI Chat Completions API, max_tokens is an early field used to limit the total number of tokens in the generated content; OpenAI introduced max_completion_tokens in the o1 series and later reasoning models to constrain the token upper limit of the entire completion, and gradually recommends that new models use the latter.
50%