{"id":585127,"date":"2025-09-08T13:56:47","date_gmt":"2025-09-08T12:56:47","guid":{"rendered":"https:\/\/blog.jetbrains.com\/?post_type=ai&p=585127"},"modified":"2025-09-08T13:56:52","modified_gmt":"2025-09-08T12:56:52","slug":"jetbrains-mellum-now-available-on-amazon-bedrock-marketplace-for-free","status":"publish","type":"ai","link":"https:\/\/blog.jetbrains.com\/zh-hans\/ai\/2025\/09\/jetbrains-mellum-now-available-on-amazon-bedrock-marketplace-for-free","title":{"rendered":"JetBrains Mellum Now Available on Amazon Bedrock Marketplace \u2013 For Free"},"content":{"rendered":"\n

Production-grade code completion at your fingertips \u2013 deploy Mellum instantly with full control and zero extra model fees<\/em><\/p>\n\n\n\n

JetBrains Mellum<\/a> \u2013 our focused large language model (LLM) for code completion \u2013 is now available on the Amazon Bedrock Marketplace. This means Mellum is easier than ever to use in real-world applications, with instant access via major cloud providers.<\/p>\n\n\n\n

Whether you’re integrating AI into your existing software engineering workflows or launching a new ML-backed service, Mellum can now be deployed in seconds \u2013 with no additional model fees and complete control over inference costs.<\/p>\n\n\n\n

How does it compare to Mellum\u2019s open source version?<\/h2>\n\n\n\n

Earlier this year, we released an open-source version of Mellum<\/a> on Hugging Face to support the developer community with an inspectable, adaptable model designed for code.<\/p>\n\n\n\n

But the version available on Amazon Bedrock Marketplace goes further.<\/p>\n\n\n\n

This is our production-grade Mellum, improved and optimized for enterprise deployment. It delivers better performance, greater stability, and is the same version we rely on internally at JetBrains for real-world workloads.<\/p>\n\n\n\n

Flexible, cost-transparent inference<\/h2>\n\n\n\n

With Amazon Bedrock Marketplace, deploying Mellum is as simple as: <\/p>\n\n\n\n

1) Selecting the model from the Bedrock marketplace <\/p>\n\n\n\n

2) Choosing the GPU hardware that matches your performance\/cost needs<\/p>\n\n\n\n

3) Hitting deploy<\/p>\n\n\n\n

You get full control over the hardware configuration and pricing, and Mellum runs without any additional licensing or model hosting fees. You only pay for the compute \u2013 and only for what you use.<\/p>\n\n\n\n

Ideal for hands-on engineering use<\/h2>\n\n\n\n

Mellum on Bedrock is perfect for teams who want:<\/p>\n\n\n\n