mammoth-1
A 27B decision model from Base Labs, Baseten's research lab, fine-tuned from Qwen3.5-27B. It answers multiple-choice, yes/no and ordinal scoring questions with probabilities, with a 16K context. MIT-licensed weights are on Hugging Face; Baseten serves it on its Model APIs and has not published a price.
No host has a verified price for mammoth-1 yet. The weights are open, so you can run it yourself.
Every decision model with its hosts and prices is on the decision models page. For text-generating models, see LLM API pricing by model.
Compare every inference provider on free tiers, hosting and EU availability in the inference APIs table.
Is your product missing?