Itura provides a robust serverless environment built on leading cloud infrastructure like AWS and Google Cloud Platform, utilizing Docker containers for secure and isolated execution of your agentic applications.

​
Supported Runtimes

Your agent code can be written using Python versions 3.9, 3.10, 3.11, 3.12, and 3.13. Itura ensures the execution environment is pre-configured with the selected Python runtime when deploying your agent.

​
LLMs, frameworks, and libraries

Itura is agnostic to the LLM or framework you use. Leverage the full power of the Python ecosystem. Whether you’re using LangChain, LlamaIndex, OpenAI’s SDK, or any other library or framework, simply include it in your requirements.txt file, and Itura will handle the installation during deployment. You are free to connect to any LLM provider.

​
Resources

Each agent invocation runs in an environment with 1 GB of RAM and 10 GB of ephemeral disk space. This storage is temporary and primarily used for the agent’s execution needs, including dependency installation and temporary file operations during the run. If you need to persist large amounts of data, consider using a database or other storage solution.

​
Execution Time

A single agent invocation can run for a maximum of 10 minutes. This limit is designed for focused, task-oriented agent operations. For longer processes, consider breaking them down into smaller, invokable tasks.

​
Concurrency

Itura is designed to support concurrent agent invocations. The number of agent invocations that can run simultaneously depends on your subscription plan. Higher tiers allow for more parallel processing, suitable for high-volume applications.

​
Code package size

The total size of your uploaded agent code, including all dependencies specified in requirements.txt, must not exceed 1 GB after decompression.

​
Logging

Any output your agent prints to standard output (stdout) or standard error (stderr) is automatically captured. These logs are crucial for debugging and monitoring agent behavior and can be easily viewed directly within the Itura UI for each execution.

​
Logs retention

The duration for which execution logs are stored and accessible varies based on your subscription plan. Higher tiers offer longer retention periods, allowing you to review historical agent activity over extended times.