We made our decision model's API free (and the weights are open)
Most "AI decisions" in production aren't open-ended generation. They're classification in disguise: which team should handle this ticket? is this e-mail phishing? what's the total on this invoice? which sentence in the contract supports that answer? Teams usually send each of these to a large LLM: 1–2 seconds and a bill per call, a JSON answer to parse, and a "confidence" that means nothing. We…
Most AI decisions in production are not open-ended generation but rather classification in disguise, such as determining which team should handle a ticket or whether an email is phishing. Teams typically send these queries to a large LLM, which takes 1-2 seconds and incurs a per-call charge, returning a JSON answer and a confidence level that holds little significance.
THX-01, a model built for such decisions, is now making its hosted API free for use without a key or sign-up. To try it, simply send a request to https://api.hal-x.ai/v1/systemone with a JSON payload containing the state of the query, the specific questions to be answered, and the criteria for each question. The free tier allows 200 decisions per minute per IP, with each question counted as one decision.
The model's responses include probabilities for each option, computed in a single forward pass of about 10 ms. THX-01 has been benchmarked against other models on a 2,843-ticket dataset, showing higher accuracy and faster latency. The model's confidence is calibrated, enabling it to handle about 75% of tickets automatically with zero errors across various test sets. The model's weights are licensed under Apache 2.0, making them freely available for use.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
