# Mammoth 1 model sizes and sources

> Mammoth 1 is Baseten's 27B open decision model for text and images. OpenRouter listed it at $0.10 per million input tokens on 30 Oct 2026.

Updated 2026-11-04 (Australia/Adelaide)

By Baseten (Base Labs) · Released: 2026-10-21 · OPEN

Two open decision models from Base Labs, Baseten's research team: [mammoth-1](<https://decisionmodels.ai/models/baseten-mammoth>) (27B) and elephant-1 (9B), post-trained from Qwen3.5. They read text and images and return a probability for every option of each question, with all questions sharing one prefill of the state. Baseten launched them claiming the top spot on the community Jev Decision Index. On 29 Oct a third, smaller model joined: joey-1 (2B, built on Qwen3.5-2B), pitched as the fastest and cheapest.

## Facts

- Sizes: [mammoth-1](<https://decisionmodels.ai/models/baseten-mammoth>) 27B (Qwen3.5-27B), elephant-1 9B (Qwen3.5-9B) and joey-1 2B (Qwen3.5-2B), 16,384-token context
- Licence: MIT
- Price: OpenRouter lists [mammoth-1](<https://decisionmodels.ai/models/baseten-mammoth>) at $0.10, elephant-1 at $0.035 and joey-1 at $0.01 per million input tokens, output free (as of 30 Oct 2026) · [Compare prices](<https://decisionmodels.ai/pricing>)

## Claims and sources

- **VENDOR-REPORTED**: Baseten says [mammoth-1](<https://decisionmodels.ai/models/baseten-mammoth>) ranked #1 of 208 models on Jev Decision Index 0.4 with a full score of 71.24, and elephant-1 is the best decision model under 15B parameters (62.65) [source](<https://labs.baseten.co/articles/mammoth-1-and-elephant-1-frontier-decision-models>)
- **INDEPENDENT**: In the index's current public data (Decision Index 0.4.1, generated 22 Oct 2026), [mammoth-1](<https://decisionmodels.ai/models/baseten-mammoth>) is first of 228 on both public-benchmark scores (75.34 skill, 81.19 raw) but third on the full score (68.44), behind [GLiDE](<https://decisionmodels.ai/models/glide>) v1.2 (71.52) and jeffan-31b (69.75) [source](<https://huggingface.co/spaces/multimodalart/jev-decision-index>)
- **VENDOR-REPORTED**: Baseten says joey-1 has a P50 latency of 20 ms, 5x faster than comparable decision models [source](<https://www.baseten.co/blog/decisions-for-a-penny-introducing-joey-1/>)

## Links

- [Source: Announcement](<https://www.baseten.co/blog/the-best-decision-models-in-the-world-mammoth-1-and-elephant-1/>)
- [Source: Technical report](<https://labs.baseten.co/articles/mammoth-1-and-elephant-1-frontier-decision-models>)
- [Source: mammoth-1 on Hugging Face](<https://huggingface.co/baselabs/mammoth-1>)
- [Source: elephant-1 on Hugging Face](<https://huggingface.co/baselabs/elephant-1>)
- [Source: joey-1](<https://huggingface.co/baselabs/joey-1>)

[JevBench scores and methodology](<https://decisionmodels.ai/jevbench>)

## Related projects

No related projects are listed in the registry.