How to use from the
Use from the
Transformers library
# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-classification", model="baselabs/mammoth-1")
# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("baselabs/mammoth-1")
model = AutoModelForMultimodalLM.from_pretrained("baselabs/mammoth-1", device_map="auto")
Quick Links

mammoth-1

mammoth-1 is a 27B parameter decision model post-trained from Qwen3.5-27B by Base Labs at Baseten. Provide the model a state (text, optionally with images) and a set of questions and it will return a calibrated probability for every option, answering all questions from one shared-prefix prefill. On tested benchmarks:

  • #1 of 208 models on the Jev Decision Index 0.4: 71.24, ahead of Perplexity Decider v1.2 (69.42), Jev 1.14 (+9.0) and Cloudflare Clef (+15.5).
  • #1 on a replica of the Artificial Analysis Decision Index: 65.9, ahead of d1-ultra (65).

For quick, cheaper decisions, see the 9B sibling elephant-1.

πŸ—’οΈ Model Details

Model Parameters Description
Qwen3.5-27B 27B General-purpose base model
mammoth-1 27B Post-trained for single-pass, calibrated decisions
elephant-1 9B Smaller sibling, fastest decisions
  • Inputs: text and images
  • Training: full fine-tune of the language model on text and image decisions; the vision encoder is frozen
  • Attention: block-bidirectional mask on full-attention layers.
  • Max sequence length: 16,384 tokens

πŸƒ How to use

Baseten Model APIs

mammoth-1 is free for one week on Baseten Model APIs, with zero data retention on by default. The free week ends 10/27/26 at 11:59 PM PT. The endpoint accepts text and image inputs.

Self-hosted

setup.sh installs the pinned serving runtime (OneJev + serve.patch). It needs one CUDA GPU and the pytorch/pytorch:2.8.0-cuda12.8-cudnn9-devel image.

hf download baselabs/mammoth-1 --local-dir mammoth-1 && bash mammoth-1/setup.sh
from qev.engine import DecisionEngine
from qev.schema import SystemOneRequest

engine = DecisionEngine("baselabs/mammoth-1", device="cuda:0")

questions = {
    "refund": {"type": "noul", "instructions": "Is the customer asking for a refund?"},
    "team": {
        "type": "choice",
        "instructions": "Which team should handle this?",
        "criteria": {"billing": "Charges, refunds, invoices",
                     "technical": "App or site faults",
                     "fraud": "Suspected unauthorised use"},
    },
    "urgency": {
        "type": "score",
        "instructions": "How urgent is this?",
        "criteria": ["Can wait", "Today", "Blocking the customer now"],
    },
}
resp, _ = engine.decide(SystemOneRequest(
    state="I was charged twice this month, please refund one of them.", questions=questions))
print(resp.answers)
type criteria answer fields
noul: yes or no optional noul: P(yes)
choice: one of named options {name: description} choice, probabilities
score: ordered levels list of levels, lowest first score, probabilities

Requests longer than the trained length limits are refused, not truncated.

πŸ“Š Performance

Jev Decision Index 0.4

Model Decision Index
mammoth-1 71.24
Perplexity Decider v1.2 69.42
elephant-1 62.65
Jev 1.14 62.2
Cloudflare Clef 55.7

On the public split, mammoth-1 leads in 4 of 5 areas (knowledge & reasoning, language, tools & automation, retrieval & classification, arts & human taste).

Artificial Analysis Decision Index (our replica)

Model Score
mammoth-1 65.9
elephant-1 65.2
d1-ultra 65

πŸ”§ Training (all details in the technical report)

  • Data: 6.21M decision examples (Mix 8). 58.3% core decision skills, 16.1% safety and abstention, 16.1% judging and expert domains, 9.5% public decision data.
  • Objective: cross-entropy plus a Brier calibration term (Ξ» = 1), with a trained "I don't know" option.
  • Recipe: 1 epoch (9,859 steps), peak LR 2.7e-6, AdamW (Ξ² 0.9/0.95, no weight decay), 3% warm-up, cosine decay, bf16.
  • Hardware: 18 nodes Γ— 4 GB300 (72 GPUs).

Citation

@article{baseten2026mammoth,
  author  = {Base Labs, Baseten},
  title   = {mammoth-1 and elephant-1: Frontier Decision Models},
  journal = {Base Labs},
  year    = {2026},
  note    = {https://labs.baseten.co/articles/mammoth-1-and-elephant-1-frontier-decision-models},
}

License

mammoth-1 is fine-tuned by Baseten from Qwen3.5-27B, which is licensed under the Apache License 2.0. Baseten's modifications are released under the MIT License. Use of the model is subject to both licenses; see LICENSE and NOTICE.

Attribution

Base model: Qwen3.5-27B, Copyright 2026 Alibaba Cloud, licensed under the Apache License 2.0. Baseten modified the weights by further training. Qwen is a trademark of Alibaba Cloud. This model is not affiliated with or endorsed by Alibaba Cloud.

Downloads last month
12,791
Safetensors
Model size
27B params
Tensor type
BF16
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for baselabs/mammoth-1

Base model

Qwen/Qwen3.5-27B
Finetuned
(355)
this model
Finetunes
24 models
Merges
1 model
Quantizations
9 models

Space using baselabs/mammoth-1 1