Endpoint Details
Review how to make this request and handle its possible responses.
https://integrate.api.nvidia.com/v1/chat/completionsInvokes inference using the model chat parameters.
Request Body
Edit the request fields below.
Request details
Parameters8 Parameters
messagesrequiredA list of messages comprising the conversation so far.
1messages[].rolerequiredThe role of the message's author.
messages[].contentrequiredThe contents of the message. <br>To pass media (only with role=`user`): <br> - Use content parts. Text is passed with `type`=`text`. <br> - Images are passed with `type`=`image_url`; set `image_url.url` to an image URL or a base64 data URI like `data:image/{format};base64,{base64encodedimage}`. <br> - Videos use `type`=`video_url`; audios use `type`=`audio_url` or `type`=`input_audio` when supported by the model schema. <br>For `system` and `assistant` roles, the object list format is not supported.
modelrequiredThe model to use.
nvidia/nemotron-3-nano-omni-30b-a3b-reasoningmax_tokensoptionalThe maximum number of tokens that can be generated.
65536reasoning_budgetoptionalMaximum number of tokens the model is allowed to use for internal reasoning before the reasoning trace is forced to end. Use `-1` to disable budget enforcement. This can also be provided via `chat_template_kwargs.reasoning_budget` for backwards compatibility.
16384Minimum: -1Maximum: 32768seedoptionalChanging the seed will produce a different response with similar characteristics. Fixing the seed will reproduce the same results if all other parameters are also kept constant.
streamoptionalIf set, partial message deltas will be sent, like in ChatGPT. Tokens will be sent as data-only [server-sent events](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events/Using_server-sent_events#Event_stream_format) as they become available, with the stream terminated by a `data: [DONE]`
falsetemperatureoptionalWhat sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
0.6top_poptionalAn alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or `temperature` but not both.
0.95Headers3 Headers
AuthorizationrequiredFormat: Bearer $NVIDIA_API_KEY
Content-TyperequiredAcceptoptional