Featherless integrates MiniMax M3.1 Flash model for coding
San Francisco-based inference platform Featherless will add the MiniMax M3.1 Flash model to its service offerings on November 4.
Serverless inference provider Featherless announced it will integrate the MiniMax M3.1 Flash model starting November 4. The model is built specifically for coding, agentic workflows, and computer-use tasks, featuring 428 billion parameters with 23 billion active per token.
In addition to its technical specifications, the model supports vision capabilities through an OpenAI-compatible API. According to the Artificial Analysis Intelligence Index, the model currently holds a score of 29.
This release follows broader availability of the model, which is already listed on platforms such as OpenRouter. By adding this model to its platform, Featherless aims to provide developers with updated tooling for automation and software development tasks.