Understand this story
On October 24, 2026, Mubit launched a platform that lets teams train their own AI models for repetitive tasks. According to a report by Rohan Paul, the service reads the API calls that agents have already sent to OpenAI or Anthropic, identifies repetitive tasks, and trains users' own small models, making this portion of the work cheaper and with shorter waits, while the remaining tasks are still handed off to frontier models; integration only requires replacing the base URL and keeping the existing client code, and it can also connect directly to already-collected traces, and the company says the models refresh every few minutes. On the same day, Testing Catalog reported that the platform sits as a proxy base URL in front of the OpenAI, Anthropic, or Gemini APIs, trains small models based on agent tasks, and in Mubit's tests can take over tasks once it catches up to the frontier models, and can also learn from historical calls imported from Langfuse, Braintrust, or Pydantic Logfire; the report also mentioned that Mubit released open-source models and a new benchmark. There are currently no subsequent developments.
AI-generated from public reports
Developments
1 confirmed developments
- Source:Rohan Paul
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Oct 24
Testing Catalog
Mubit launches a platform for teams to train their own small models, and releases an open-source model and a new benchmark
Mubit launches a platform that lets teams train their own AI models for repetitive tasks, while also releasing an open-source model and a new benchmark. The platform sits as an agent base URL in front of OpenAI, Anthropic, or Gemini APIs, trains small models based on agent tasks, and in Mubit's tests can take over tasks once it catches up to frontier models. The platform can also learn from historical calls imported from Langfuse, Braintrust, or Pydantic Logfire.
Rohan Paul
Mubit launches a service that trains your own small model using agent API calls
Mubit reads the API calls that agents have already sent to OpenAI or Anthropic, identifies repetitive tasks, and trains a user's own small model so that this portion of the work costs less and has shorter wait times, while the remaining tasks are still handled by frontier models. Integration only requires replacing the base URL and keeping the existing client code; it can also connect directly to traces that have already been collected. The company says the model refreshes every few minutes. The author believes the idea is to write experience into weights rather than prompts, so agents from different companies will train different small models.
Heat trend for this story
- Current
- 5
- Period peak
- 6 Oct 25 22:00
- 24-hour change
- —
Move the pointer or select the chart to inspect hourly heat. Use the left and right arrow keys with a keyboard.