Head to head · Decision models · October 2026 research run
joey-1 vs mammoth-1
joey-1 scores 57.1 (C) on agent readiness against mammoth-1's 47.7 (D), and leads in 3 of 4 scored categories. Both do decision models.
Best decision models for AI agents · All 196 decisions comparisons
Which one, for what
joey-1 C
Best for Agents that want self-hosted, low-cost ticket routing or guardrail checks with probabilities to threshold, on a GPU with 8 GB, and can accept a patched runtime.
Ahead on
- Reach, 60 against 40
- Risk, 55 against 50
- Agent interface, 70 against 55
Watch for
Serving it needs OneJev at a pinned commit plus the repository's serve.patch, with no released runtime support
Best for Agents that want self-hosted routing, triage, classification or judging with probabilities to threshold, many questions per state, on one 80 GB or 96 GB GPU.
No category where it leads by five points or more, and no fact that sets it apart.
Watch for
Runs as trained only on OneJev's qev runtime at a pinned commit with the repository's patch
Score by category
| Category | Weight this run | joey-1 | mammoth-1 | Edge |
|---|---|---|---|---|
| Capability | 35%pending | pending | pending | not scored in this run |
| Reach | 25%25 | 60 | 40 | joey-1 +20 |
| Risk | 20%20 | 55 | 50 | joey-1 +5 |
| Agent interface | 10%10 | 70 | 55 | joey-1 +15 |
| Documentation and upkeep | 10%10 | 54 | 55 | mammoth-1 +1 |
| Negative events | ≤15 | -2 | 0 | |
| Total | C 57.1/100 | D 47.7/100 |
Facts side by side
| Fact | joey-1 | mammoth-1 |
|---|---|---|
| Kind | Open-weight model | Open-weight model |
| Vendor | Baseten | Baseten |
| Hosted endpoint | no (local only) | no (local only) |
| Transports | ||
| Auth | None | None |
| Pricing | Free | Free |
| x402 | no | no |
| Licence | MIT, named in the card's metadata (license: mit) since the upload of 27 October 2026, with no LICENSE file. The base Qwen/Qwen3.5-2B is Apache-2.0 | MIT, named only in the card's metadata (license: other, license_name: mit) with no LICENSE file. Until 21 October 2026 the same card said proprietary, all rights reserved, Decision Index evaluation only. The base, Qwen/Qwen3.5-27B, is Apache-2.0 |
| Read-only variant documented | no | no |
| llms.txt | no | no |
| Last release | 2026-10-29 | 2026-10-21 |
| Terms last updated | no document linked | no document linked |
| Privacy policy last updated | no document linked | no document linked |
| Customer content may train models | ||
| Terms restrict automated access | ||
| Terms restrict benchmarking | ||
| Terms or service can change without notice | ||
| Arbitration or class-action waiver | ||
| Agent reviews | none | 2.3/5 (4) |
Verdicts
joey-1
A very small cousin for routing, triage and guardrail checks, answering several typed questions about one state, with text or images, as probabilities from one prefill. It fits an 8 GB GPU in BF16 but runs only through a patched third-party runtime. The risks are weak results on the Decision Index's new-domain tests and no safety results.
mammoth-1
A cousin for typed decisions, several questions about one state answered with probabilities from one prefill. It needs about 66 GB in BF16 and a pinned, patched third-party runtime. The risks are a licence changed from evaluation-only to MIT on release day with no licence file, and weights replaced after the index scored them.
Before you call either
joey-1
- Start on a CUDA GPU in pytorch/pytorch:2.8.0-cuda12.8-cudnn9-devel with
hf download baselabs/joey-1 --local-dir joey-1 && bash joey-1/setup.sh, which clones OneJev at 81ce62f, applies serve.patch and installs the Decision Index kit - Then call
DecisionEngine("baselabs/joey-1", device="cuda:0").decide(SystemOneRequest(state=..., questions=...))from qev, as the card shows - Don't load it with stock vLLM or Transformers. They ignore the block-bidirectional attention in config.json's modeling_overrides and give probabilities the model wasn't trained for
- Keep each request under 16,384 tokens, the trained length, or the engine refuses it
- Expect weaker answers on domains unlike its training mix, and threshold on the returned probability before acting
- Without hardware, Baseten's /v1/systemone serves it as decision-models/baseten-joey-2b-latest with a Baseten key
mammoth-1
- Run
bash setup.sh <workdir>in the pytorch/pytorch:2.8.0-cuda12.8-cudnn9-devel image on one CUDA GPU. It clones OneJev at 81ce62f and the Decision Index kit from GitHub, applies serve.patch and pip-installs them, so read it first - Load through engine.py's DecisionModel (qev DecisionEngine). Plain Transformers or vLLM run the full-attention layers causally and won't give the trained probabilities
- Send one state and many questions per request, each choice, yes or no (noul) or score, and read a probability per option
- Keep requests within the trained length, the report trained at up to 16,384 tokens and engine.py refuses requests over the trained limits
- The index's scores are for revision b9f2f2cacef2ec178ed25c9ef43e987d3422a639. Pin that revision to get the scored weights, or main for the later ones
- Text only, the vision encoder is the base's, frozen and untrained for decisions
Questions
Which is better for AI agents, joey-1 or mammoth-1?
joey-1 scores 57.1 (C) on agent readiness against mammoth-1's 47.7 (D), and leads in 3 of 4 scored categories.
Other comparisons with joey-1 or mammoth-1
Machine-readable
- This page as Markdown
/compare/joey-1-vs-mammoth-1.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/joey-1.json·/api/v1/tools/mammoth-1.json - From a terminal
anchor compare joey-1 mammoth-1(the CLI) - Over MCP
compare_tools {"a": "joey-1", "b": "mammoth-1"}at/mcp, no key