
    Amazon Bedrock now offers on-demand deployment for custom Meta Llama models, allowing customers to reduce costs by processing requests in real time without pre-provisioned compute resources.


<div>
<p>
Amazon Bedrock now offers on-demand deployment for custom Meta Llama 3.3 models, providing a cost-effective solution for AI model usage.
</p>
<ul>
<li>Available for models fine-tuned or distilled in Bedrock after September 15, 2025</li>
<li>Enables real-time request processing without pre-provisioned compute resources</li>
<li>Pay-per-use model eliminates need for always-on infrastructure</li>
<li>Part of Amazon Bedrock's fully managed service offering</li>
<li>Provides secure and responsible AI application development</li>
</ul>
<p>
This feature allows customers to optimize costs and flexibility when working with custom AI models.
</p>
</div>


Announcing on-demand deployment for custom Meta Llama models in Amazon Bedrock

Related articles

Related articles

Apr 23
2024
Meta Llama 3 models are now available in Amazon Bedrock

Apr 29
2025
Meta’s Llama 4 now available fully managed in Amazon Bedrock

Apr 29
2025
Llama 4 models from Meta now available in Amazon Bedrock serverless

Jul 23
2024
Meta Llama 3.1 generative AI models now available in Amazon Bedrock