infery
← All models

Nemotron Diffusion Vlm

nemotron-diffusion-vlm

Visionby NVIDIA

Nemotron-Labs-Diffusion-VLM-8B is the vision-language extension of the Nemotron-Labs-Diffusion family.

Details

Accepts
image

Pricing

Input
625 cr / 1M tokens

Prices in credits (1 credit = $0.01).

Data schema

Input

FieldTypeDescription
promptrequiredstringPrompt to answer about the image.
image_urlrequiredstringURL of the image to be processed.
thresholdnumberConfidence threshold used by diffusion decoding.
max_tokensintegerMaximum number of tokens to generate.
block_lengthintegerBlock length used by diffusion decoding.
num_inference_stepsNumber of diffusion decoding steps. Defaults to 256, rounded up only when omitted and required by the upstream block schedule. Explicit values must be at least max_tokens / block_length and divisible by max_tokens / block_length.

Output

FieldTypeDescription
usageToken and diffusion decoding usage information.
outputstringGenerated answer.
timingsRequest timing breakdown in seconds.