# Drex 1.5 (slim)

> Drex 1.5 is Nace.AI's decision model. It returns a probability for each option in typed yes or no, choice and score questions. Hosted at console.nace.ai, with open weights under the Nace AI Open RAIL-M licence.

- Full: https://www.anchorterminal.com/tools/nace-drex.md (~7,300 tokens) · this version ~2,130 tokens · JSON https://www.anchorterminal.com/tools/nace-drex.json · canonical https://www.anchorterminal.com/tools/nace-drex
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-10

**B · 69.7/100 · rank #172 of 954 · #2 in Decision models · not agent-ready · confidence medium**


Assessment: The hosted endpoint has a public OpenAPI document, published limits and pinned model versions, at $0.05 per million input tokens with no output charge. The #1 under 10B claim rests on Nace's own run, since Drex is absent from both public boards. Weights carry Open RAIL-M, which requires a commercial licence above $1 million in revenue or funding.

## Facts

- Kind: Model API · vendor: Nace AI, Inc. (Nace.AI) · category: Decision models · legal entity: Nace AI, Inc. · provenance 75/100
- Endpoint: `https://console.nace.ai/v1/systemone` (HTTP)
- Auth: API key · pricing: Freemium · x402: no · licence: Hosted API under the Nace AI Terms of Service, last updated 5 May 2026. Weights under the Nace AI Open RAIL-M licence (modified, version 1.0, October 2026), which is not an open-source licence. Businesses above $1 million in revenue or funding need a separate commercial licence, and competing products are excluded. The bundled code/kev package, derived from Jared Palmer's Kev (https://github.com/jaredpalmer/kev), and the GitHub repository are Apache-2.0, and the agent skill is MIT.
- Probe metrics: not measured yet (probes haven't run)
- Surface graded: The hosted Drex API at console.nace.ai, with POST /v1/systemone for decisions and GET /v1/models for the model list. The open weights and the bundled code/kev server, derived from Kev (https://github.com/jaredpalmer/kev), are noted but not graded, since a self-hosted server is software the owner runs.
- Models: drex-v1.0, released 24 September 2026 at $0.04 per million input tokens with a 32,768-token state limit. drex-v1.5, released 28 September 2026 at $0.05 with a 131,072-token state limit. The drex-latest alias points to drex-v1.5. drex-v1.1 is retired and answered by drex-v1.5.
- Request limits: 512 questions per request, a 1,048,576-byte body and a 55-second upstream timeout. A state over the token limit returns 422.
- Rate limits: Free tier 120 requests a minute with 8 in flight. Paid tier, from the first completed top-up, 600 a minute with 16 in flight. A 429 carries retry-after and retry-after-ms. A 529 means the model service did not answer within 55 seconds.
- Credit: Signup credit of $25 expires after three months. Monthly credit of $5 expires at the next renewal. Top-ups from $5 to $1,000 by card expire after 12 months. Only a 200 response is debited, once per request_id.
- Weights: 8.95 billion parameters in bf16 across four safetensors shards, about 18 GB. The Q8_0 GGUF is about 9.5 GB. The base model is XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B with a pointer head.
- Self-hosted context: The bundled server accepts 16,384 tokens by default and up to 131,072 when KEV_CONTEXT or SYSTEMONE_CONTEXT is raised. The llama.cpp route needs -c 262144 for the full length.
- Benchmarks: Vendor claims only. Nace reports 58.08 on Decision Index 0.3.1 (public only), 0.12 points above Jev 1.13.0 at 57.96, inside the board's 0.9-point tie band. The site reports 58.28 on Decision Index 0.2.1 and says Drex 1.5 leads all 71 entries under 10B parameters. Neither public edition lists Drex, and Jev's parameter count is unpublished.
- SDKs and MCP: Python nace-sdk 0.1.0 on PyPI and TypeScript nace-sdk 0.1.0 on npm (Apache-2.0), both published on 8 October 2026. The nace-mcp 0.1.0 server on PyPI has 21 tools, of which decide and list_models cover decisions. A Claude Code plugin and an agent skill (MIT) cover Claude Code, Codex, Cursor, OpenCode, Hermes Agent and OpenClaw.
- Downloads: npm nace-sdk had 318 downloads in the week to 8 October 2026. PyPI recorded 156 downloads of nace-sdk and 118 of nace-mcp in the last seven days. Hugging Face drex-v1.5 shows 37 downloads and 16 likes on 10 October 2026.
- Hosting and enterprise tiers: The Trust Centre names AWS and Azure as infrastructure. No regional routing or residency commitment for Drex was found. Dedicated Cloud, Your Cloud and On-Prem tiers are sales-led on the developer page.
- Trust: SOC 2 Type 2 is listed on trust.nace.ai, but the report and the pentest report sit behind an access request and were not read.
- Prices: Drex 1.5 input $0.05 per 1M tokens; Drex 1.5 output free per 1M tokens

- Scores: Reliability 80, Performance pending, Schema & documentation 88, Agent ergonomics 85, Security & auth 65, Payments & pricing 40, Task success pending, Maintenance & community 65, Transparency & trust 63 · negative events -2 · total over the 7 assessed categories
- Why: Reliability, A public status page with component history, a BetterStack page, names an API and dashboard component that serves /v1/systemone, and it show… · Schema & documentation, A public OpenAPI 3.1 document for the Drex API at console.nace.ai/openapi.json, with POST /v1/systemone and its request schema (25). · Agent ergonomics, A decision answer is one probability per option, and no tokens are generated, so the response is compact (25). · Security & auth, Bearer keys in the nace_sk_ format are revocable and can be rotated without downtime, and an account is capped at three active keys. · Payments & pricing, No machine payment protocol on the API, and no x402, MPP or L402 in the pricing reference, the API reference or the OpenAPI document (0). · Maintenance & community, The last model change was on 28 September 2026, 12 days before this check, which earns the full 30. The docs list dated model entries in the… · Transparency & trust, The hosted service has a named licensor and terms with a stated forum, and the weights carry a published licence text.
- Sources: 25, open questions: 10, both in the full twin
- Capabilities: inference.decision
- JSON: https://www.anchorterminal.com/api/v1/tools/nace-drex.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/nace-drex.svg` or a link to https://www.anchorterminal.com/tools/nace-drex from a page on nace.ai or one of its subdomains, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Call https://console.nace.ai/v1/systemone with Authorization: Bearer nace_sk_..., and pin model drex-v1.5 rather than drex-latest in production.
2. Stop on 402 insufficient_credit or 402 payment_required until credit is topped up or the unpaid invoice is paid.
3. Retry 429 and 529 after retry-after-ms. On a 422 for an unknown model, use one of the ids the error message lists.
4. Keep state below 131,072 tokens and the body below 1,048,576 bytes, and send at most 512 questions per request.
5. Keep the key on a server. A key in browser code is exposed, and the /v1 routes send no Access-Control-Allow-Origin header.
6. Check usage.input_tokens and the usage CSV after retries, since /v1/systemone documents no idempotency key.

## Connect

```bash
pip install nace-sdk   # or: npm install nace-sdk
```

```bash
curl https://console.nace.ai/v1/systemone \
  -H "Authorization: Bearer $NACE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"drex-v1.5","state":"I was charged twice for order A-104.","questions":{"refund":{"type":"noul","instructions":"Does the customer ask for a refund?"}}}'
```

```bash
uvx nace-mcp login
claude mcp add nace -- uvx nace-mcp
```

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| OpenAI Decisions API | BB | 71.5 | inference.decision | https://www.anchorterminal.com/tools/openai-decisions-api.min.md |
| Decider | B | 69.5 | inference.decision | https://www.anchorterminal.com/tools/decider.min.md |
| Laya | B | 69.2 | inference.decision | https://www.anchorterminal.com/tools/convai-laya.min.md |
| Kev | B | 67.4 | inference.decision | https://www.anchorterminal.com/tools/jaredpalmer-kev.min.md |
| Vela 2.0 | B | 66.5 | inference.decision | https://www.anchorterminal.com/tools/vela.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)