# Mammoth-1

Decision models · checked 2026-10-22 · https://dbggr.dev/models/mammoth-1/

Open-weight 27B decision model on Qwen3.5-27B for classification, routing and scoring

Mammoth-1 is an open-weight decision model from Baseten (Base Labs) (MIT, 27B parameters): it answers typed questions, such as yes or no, pick one or pick several, with probabilities instead of writing text. It's currently free to use (Baseten Model APIs, Vercel AI Gateway). JevBench, the main independent benchmark, doesn't list it yet.

- Made by: Baseten (Base Labs)
- Weights: Open, MIT
- Size: 27B
- Where: Baseten Model APIs, Vercel AI Gateway (both free until 27 Oct)
- Images: No, text only
- Context: 16,384 (max sequence length)
- Fine-tuning: Not stated
- Docs: https://huggingface.co/baselabs/mammoth-1

## Mammoth-1 price

US dollars per million input tokens. Decision models don't write text, so none of the hosted ones charge for output.

| Where | Input | Output | Notes | Checked |
|---|---|---|---|---|
| Baseten Model APIs, Vercel AI Gateway (both free until 27 Oct) | Free | Free | Free on Baseten's Model APIs and Vercel's AI Gateway through 27 October 2026 (11:59 PM PT). Baseten's pricing page lists no per-token price yet. | [2026-10-22](https://vercel.com/changelog/basetens-mammoth-1-and-elephant-1-now-available-on-ai-gateway-for-free) |

## Decision models like Mammoth-1

| Model | Input / M | JevBench | Weights |
|---|---|---|---|
| Mammoth-1 | Free | – | Open |
| [d1-ultra](https://dbggr.dev/models/d1-ultra/) | $0.042 | 80.5 | Closed |
| [Jev 1.13](https://dbggr.dev/models/jev/) | $0.042 | 77.1 | Closed |
| [Clef](https://dbggr.dev/models/clef/) | $0.24 | 75.3 | Open |
| [d1](https://dbggr.dev/models/d1/) | $0.02 | 74.4 | Closed |
| [OpenAI Decisions (GPT-6 Luna)](https://dbggr.dev/models/openai-decisions/) | $0.10 | 73.5 | Closed |
| [Microsoft-Decision-1](https://dbggr.dev/models/microsoft-decision-1/) | $0.042 | 70.8 | Closed |
| [Clef-flash](https://dbggr.dev/models/clef-flash/) | $0.038 | 63.8 | Open |
| [Clef-omni](https://dbggr.dev/models/clef-omni/) | $0.15 | 58.3 | Open |
| [d1-3B](https://dbggr.dev/models/d1-3b/) | Self-host | 49.5 | Open |
| [d1-omni-600M (experimental)](https://dbggr.dev/models/d1-omni-600m/) | Self-host | 32.8 | Open |
| [Laya](https://dbggr.dev/models/laya/) | Free | 32.5 | Open |
| [CLM-8B](https://dbggr.dev/models/clm-8b/) | Self-host | 20.1 | Open |
| [Clairic](https://dbggr.dev/models/clairic/) | $0.10 | – | Closed |
| [Elephant-1](https://dbggr.dev/models/elephant-1/) | Free | – | Open |
| [Strands Decider 2B](https://dbggr.dev/models/strands-decider-2b/) | Self-host | – | Open |

## FAQ

### How much does Mammoth-1 cost?

Mammoth-1 costs nothing at Baseten Model APIs, Vercel AI Gateway (both free until 27 Oct). Free on Baseten's Model APIs and Vercel's AI Gateway through 27 October 2026 (11:59 PM PT). Baseten's pricing page lists no per-token price yet. Checked 22 October 2026.

### Is Mammoth-1 open weight?

Yes. Baseten (Base Labs) publishes the weights under MIT, so you can run and fine-tune it yourself.

### How fast is Mammoth-1?

Baseten (Base Labs) doesn't publish a latency figure.

### How does Mammoth-1 compare with Jev?

JevBench doesn't list Mammoth-1 yet, so there's no independent comparison. Jev costs $0.042 per million input tokens; Mammoth-1 costs Free.