nace-ai/

drex-v1.5

$0.04 / 1M tokens

Drex v1.5 is a 9B decision model from Nace.AI. Given a state (text or JSON) and typed questions (yes/no, multiple choice, or ordinal score), it returns a calibrated probability for every option in a single forward pass, with no text generation. It is TypeSafe System One compatible, handles contexts up to 131K tokens, and scores 58.28 on Decision Index 0.2.1, ahead of Jev 1.13.0.

Public

Zero retention

131,072

Project License

api

Log in to run decisions.

Model Information

Drex v1.5

Drex v1.5 is a decision model from Nace.AI. You give it a state (text or JSON) and typed questions, and it returns a calibrated probability for every option in a single forward pass. It generates no text, and it never answers with an option you did not supply.

About 9B parameters, built on MiMo-V2.6-Distill-Qwen-9B (hybrid attention), with a trained pointer head.
Context up to 131,072 tokens for the state plus each question.
TypeSafe System One compatible: the same request and response as Jev, so the TypeSafe SDK works by changing its base URL.

Question types

Type	You send	You get back
`noul`	a yes/no question, with optional `criteria.true` / `criteria.false`	`noul`, the probability of "yes"
`choice`	options as `criteria` (name → description or `null`), 1-255 options	`choice` (the most likely option), `probabilities` per option, and `confidence`
`score`	ordered levels as a `criteria` list, 1-255 levels	`score` (the expected level), `probabilities` per level, `legend`, and `confidence`

One request can ask many questions about the same state. The state is read once and billed once per request.

Example

curl https://api.deepinfra.com/v1/decisions \
  -H "Authorization: Bearer $DEEPINFRA_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nace-ai/drex-v1.5",
    "state": "The policy allows a refund within 30 days when the customer has a receipt. The customer bought the product 12 days ago and has the receipt.",
    "questions": {
      "refund_allowed": {"type": "noul", "instructions": "Does the policy allow this refund?"},
      "next_action": {"type": "choice", "instructions": "What should the support agent do?",
                      "criteria": {"refund": "Approve the refund.", "deny": "Deny the refund.", "escalate": "Escalate for policy review."}},
      "urgency": {"type": "score", "instructions": "How urgent is this request?",
                  "criteria": ["Routine", "Time-sensitive", "Immediate"]}
    }
  }'
copy

The same API is served at https://api.deepinfra.com/v1/systemone. To use the TypeSafe Python SDK, set TYPESAFE_BASE_URL=https://api.deepinfra.com/typesafe and model="nace-ai/drex-v1.5".

Performance

Figures reported by Nace.AI on the model card.

Benchmark	Drex v1.5	Jev 1.13.0
Decision Index 0.2.1 (chance-corrected)	58.28	57.91
JevBench (231 public items)	86.2%	87.0%
Game arena vs Jev (256 games)	56.8% score (122 W / 47 D / 87 L)	—
Long documents, 32k-128k tokens	93.4% (78% when truncated to 8k)	—

License and use restrictions

Drex v1.5 is © Nace.AI, Inc. and is provided under the Nace AI Open RAIL-M License (modified). By using this model you agree to its use-based restrictions (Attachment A), including:

No fully automated decisions without meaningful human review in employment, credit, housing, insurance, healthcare, education, immigration or law enforcement.
No discrimination based on legally protected characteristics.
No medical, legal or financial advice or decisions without qualified human oversight.
No military, weapons, mass-surveillance, fraud, scam or spam use.

Read the full license before use. Drex and Nace are trademarks of Nace.AI, Inc.