Sampling parameters shape the token generation process of the model. You may send any parameters from the following list, as well as others, to OpenRouter.
OpenRouter will default to the values listed below if certain parameters are absent from your request (for example, temperature to 1.0). We will also transmit some provider-specific parameters, such as safe_prompt for Mistral or raw_mode for Hyperbolic directly to the respective providers if specified.
Please refer to the model’s provider section to confirm which parameters are supported. For detailed guidance on managing provider-specific parameters, click here.
Key: temperature
Optional, float, 0.0 to 2.0
Default: 1.0
Explainer Video: Watch
This setting influences the variety in the model’s responses. Lower values lead to more predictable and typical responses, while higher values encourage more diverse and less common responses. At 0, the model always gives the same response for a given input.
Key: top_p
Optional, float, 0.0 to 1.0
Default: 1.0
Explainer Video: Watch
This setting limits the model’s choices to a percentage of likely tokens: only the top tokens whose probabilities add up to P. A lower value makes the model’s responses more predictable, while the default setting allows for a full range of token choices. Think of it like a dynamic Top-K.
Key: top_k
Optional, integer, 0 or above
Default: 0
Explainer Video: Watch
This limits the model’s choice of tokens at each step, making it choose from a smaller set. A value of 1 means the model will always pick the most likely next token, leading to predictable results. By default this setting is disabled, making the model to consider all choices.
Key: frequency_penalty
Optional, float, -2.0 to 2.0
Default: 0.0
Explainer Video: Watch
This setting aims to control the repetition of tokens based on how often they appear in the input. It tries to use less frequently those tokens that appear more in the input, proportional to how frequently they occur. Token penalty scales with the number of occurrences. Negative values will encourage token reuse.
Key: presence_penalty
Optional, float, -2.0 to 2.0
Default: 0.0
Explainer Video: Watch
Adjusts how often the model repeats specific tokens already used in the input. Higher values make such repetition less likely, while negative values do the opposite. Token penalty does not scale with the number of occurrences. Negative values will encourage token reuse.
Key: repetition_penalty
Optional, float, 0.0 to 2.0
Default: 1.0
Explainer Video: Watch
Helps to reduce the repetition of tokens from the input. A higher value makes the model less likely to repeat tokens, but too high a value can make the output less coherent (often with run-on sentences that lack small words). Token penalty scales based on original token’s probability.
Key: min_p
Optional, float, 0.0 to 1.0
Default: 0.0
Represents the minimum probability for a token to be considered, relative to the probability of the most likely token. (The value changes depending on the confidence level of the most probable token.) If your Min-P is set to 0.1, that means it will only allow for tokens that are at least 1/10th as probable as the best possible option.
Key: top_a
Optional, float, 0.0 to 1.0
Default: 0.0
Consider only the top tokens with “sufficiently high” probabilities based on the probability of the most likely token. Think of it like a dynamic Top-P. A lower Top-A value focuses the choices based on the highest probability token but with a narrower scope. A higher Top-A value does not necessarily affect the creativity of the output, but rather refines the filtering process based on the maximum probability.
Key: seed
Optional, integer
If specified, the inferencing will sample deterministically, such that repeated requests with the same seed and parameters should return the same result. Determinism is not guaranteed for some models.
Key: max_tokens
Optional, integer, 1 or above
This sets the upper limit for the number of tokens the model can generate in response. It won’t produce more than this limit. The maximum value is the context length minus the prompt length.
Key: logit_bias
Optional, map
Accepts a JSON object that maps tokens (specified by their token ID in the tokenizer) to an associated bias value from -100 to 100. Mathematically, the bias is added to the logits generated by the model prior to sampling. The exact effect will vary per model, but values between -1 and 1 should decrease or increase likelihood of selection; values like -100 or 100 should result in a ban or exclusive selection of the relevant token.
Key: logprobs
Optional, boolean
Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned.
Key: top_logprobs
Optional, integer
An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to true if this parameter is used.
Key: response_format
Optional, map
Forces the model to produce specific output format. Setting to { "type": "json_object" } enables JSON mode, which guarantees the message the model generates is valid JSON.
Note: when using JSON mode, you should also instruct the model to produce JSON yourself via a system or user message.
Key: structured_outputs
Optional, boolean
If the model can return structured outputs using response_format json_schema.
Key: stop
Optional, array
Stop generation immediately if the model encounter any token specified in the stop array.
Key: tools
Optional, array
Tool calling parameter, following OpenAI’s tool calling request shape. For non-OpenAI providers, it will be transformed accordingly. Click here to learn more about tool calling
Key: tool_choice
Optional, array
Controls which (if any) tool is called by the model. ‘none’ means the model will not call any tool and instead generates a message. ‘auto’ means the model can pick between generating a message or calling one or more tools. ‘required’ means the model must call one or more tools. Specifying a particular tool via {"type": "function", "function": {"name": "my_function"}} forces the model to call that tool.
Key: parallel_tool_calls
Optional, boolean
Default: true
Whether to enable parallel function calling during tool use. If true, the model can call multiple functions simultaneously. If false, functions will be called sequentially. Only applies when tools are provided.
Key: verbosity
Optional, enum (low, medium, high)
Default: medium
Controls the verbosity and length of the model response. Lower values produce more concise responses, while higher values produce more detailed and comprehensive responses.