Compare image classification models and pricing

Name the labels. Get the best match first.

Find zero-shot image classification models that score how well each label you name describes an image, with no training. Send an image and 2 to 100 candidate labels; the results come back best first.

2 image classification models are listed today. All of them use one OneInfer API key and pay-as-you-go wallet.

Choose and call a zero-shot image classifier in 4 steps

Step 1 of 4

The list ranks them by their price per second of compute. Different billing units are never mixed, so the order is fair.

Lowest comparable price: OpenAI: CLIP ViT-B/32

Currently the lowest-priced option in the provider price per second of compute group. Different billing units are never mixed.

Available image classification models

Comparable models are ordered by their current normalized price. Pricing content refreshed November 9, 2026.

2 models

Available image classification models, ordered by current normalized price
Model

deepinfra

CLIP ViT-B/32 is OpenAI's image and text model: it scores how well each label you name describes an image, with no training (zero-shot).

Price
$0.00055 / second
Context
N/A

deepinfra

CLIP ViT-L/14 at 336 pixels is OpenAI's larger CLIP: it scores how well each label you name describes an image, with no training (zero-shot).

Price
$0.00055 / second
Context
N/A

How to choose

  • Image tagging
  • Product categorization
  • Content filtering
  • Photo organization

Pricing methodology

OneInfer compares only positive prices with the same billing unit. Per-minute, per-character, per-token, per-image, per-video, and per-second rates remain separate. Prices can change, so the current model page and console remain the source of truth.

Frequently asked questions

How are image classification prices compared?

These models are priced per second of compute, not per image, and compared by that rate. A call takes 30–70 ms when the model is warm and a second or more when it is not; usage.billed_seconds in each reply is the time charged.

How do I call an image classification model?

Send POST /v1/ula/classify-image with the model id, an image (a public https URL or a base64 data URL) and 2 to 100 distinct labels. Every label comes back with a score, best first.

Do I need to train the model on my labels?

No. These models are zero-shot: name any labels in the request, such as “a cat” or “a product photo”, and each is scored against the image as it is.