#data retention with llms

4 messages · Page 1 of 1 (latest)

keen hare
#

When using the API, Is there a way to control when and if the LLMs use our data for training?

unique gateBOT
#

I'm an AI assistant - my answers may be incorrect and I cannot see screenshots. Please verify information with the OpenRouter docs.

For models served by most OpenRouter providers, you have control over data retention and training:

  • Most major providers (e.g., Nebius AI Studio, DeepInfra, Together, Cloudflare, Lambda, NovitaAI, inference.net) explicitly state:
    • Prompt Training: false
    • Prompt Logging: Zero retention
  • This means your data is not used for training and is not retained for logging by default.

However, policies may differ depending on the specific provider or model. Always review the "Data Policy" for the exact model/provider you are using. You can see these details on each model's page at OpenRouter Models.

For further details about provider routing and data retention, see Provider Routing.

Quote: “Prompt Training: false. Prompt Logging: Zero retention.” Llama Guard 3 8B

If you need model- or provider-specific confirmation, please specify the exact model identifier or provider name.

-# This is an automated response powered by Answer HQ. Use !mute to stop the bot from responding.

Author Not Found

The author you are looking for could not be found.

Author Not Found

The author you are looking for could not be found.

Llama Guard 3 8B - API, Providers, Stats

Llama Guard 3 is a Llama-3.1-8B pretrained model, fine-tuned for content safety classification. Run Llama Guard 3 8B with API

abstract lotus
#

or you can do it account-wide