---
id: "baseten/mammoth-1"
also: ["baseten/mammoth-1-20261021", "baselabs/mammoth-1", "vercel/baseten/mammoth-1"]
card_usd_per_m: "0.1 in, 0.0 out, at baseten"
cheapest_list_usd_per_m: "0.1 in, 0.0 out, at baseten on openrouter"
cheapest_effective_usd_per_m: "0.0997 in, 0.0 out, at baseten on openrouter"
context: 16384
max_output: 14745
input: ["text", "image"]
output: ["decisions"]
released: "2026-10-21"
api: ["/api/alpha/decisions"]
params: ["max_tokens", "stop", "temperature"]
endpoints: 2
gateways: "openrouter (1), vercel (1)"
no_training: 2
trains: 0
retains_prompts: 0
---
# Baseten: Mammoth 1

Mammoth-1 is a 27B decision model from Baseten's Base Labs, fine-tuned from Qwen3.5-27B for classification, routing, and scoring. It answers multiple-choice, yes/no, and ordered scoring questions with probabilities over predefined options rather than generated text. Shared-prefix caching reuses context across questions, while per-option probabilities support confidence thresholds and escalation decisions.