Baseten Releases joey-1, a 2B Decision Model at $0.01 per 1M Tokens
Baseten released joey-1, a 2B open-weight decision model returning calibrated probabilities instead of text, with 20 ms P50 latency and $0.01 per million input tokens.
Key Takeaways
- Key Highlight：Baseten released joey-1, a 2B open-weight decision model returning calibrated probabilities instead of text, with 20 ms P50 latency and $0.01 per million input tokens.
- Key Topics：Baseten, Releases, Decision, Model
Baseten has released joey-1, a 2-billion-parameter open-weight model designed for decision-making. Unlike conventional language models that generate text, joey-1 outputs calibrated probabilities, making it suited for classification and routing tasks.
The model is built for low-latency production use. It reports a P50 latency of 20 milliseconds and costs $0.01 per million input tokens, positioning it as a low-cost option for real-time inference.
Because joey-1 is open-weight, developers can fine-tune or self-host it. This could broaden access to fast probabilistic models for AI applications where speed and price matter more than verbose text generation.
This page provides an editorial summary based on publicly available information. It is not a republished article. Use the source link below for the original report.