Head to head · Decision models · October 2026 research run

GLiClass vs Drex 1.5

Drex 1.5 scores 69.7 (B) on agent readiness against GLiClass's 49.9 (D), and leads in 5 of 7 scored categories. GLiClass leads on payments & pricing. Both do decision models.

Best decision models for AI agents · All 91 decisions comparisons

Which one, for what

GLiClass D

Best for Topic, intent and sentiment routing over a known label set on the owner's own CPU or GPU, where many labels must be scored at once.

Ahead on

  • Payments & pricing, 60 against 40

Also in its favour

  • No key needed to call it
  • Open source

Watch for

No calibration evidence. The cards report F1 only, and the docs tell users to calibrate thresholds on their own traffic

Drex 1.5 B

Best for Routing, triage and yes or no gates inside an agent, where the caller supplies the options and acts on the probabilities. Suited to hosted use at a low per-token price, or to open weights where the licence fits.

Ahead on

  • Reliability, 80 against 43
  • Schema & documentation, 88 against 49
  • Agent ergonomics, 85 against 60
  • Security & auth, 65 against 38
  • Maintenance & community, 65 against 44

Also in its favour

  • A hosted endpoint, with nothing to install

Watch for

The rank 1 claim comes from Nace's own run on Decision Index 0.3.1, and Drex is not on the public board for 0.3 or 0.2.1.

Score by category

CategoryWeight this runGLiClassDrex 1.5Edge
Reliability16%204380Drex 1.5 +37
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.24988Drex 1.5 +39
Agent ergonomics13%16.26085Drex 1.5 +25
Security & auth14%17.53865Drex 1.5 +27
Payments & pricing10%12.56040GLiClass +20
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.84465Drex 1.5 +21
Transparency & trust7%8.86463GLiClass +1
Negative events≤150-2
TotalD 49.9/100B 69.7/100

Facts side by side

FactGLiClassDrex 1.5
KindModel APIModel API
VendorKnowledgatorNace AI, Inc. (Nace.AI)
Hosted endpointno (local only)https://console.nace.ai/v1/systemone
TransportsHTTPHTTP
AuthNoneAPI key
PricingFreeFreemium
x402nono
LicenceApache-2.0 (library and the model weights we checked)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.
Read-only variant documentednono
llms.txtnoyes
Last release2026-07-212026-09-28
Terms last updatedno document linked
Privacy policy last updatedno 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
Popularity555 stars, 13k PyPI/wk318 npm/wk, 156 PyPI/wk

Verdicts

GLiClass

An Apache-2.0 classifier that scores a whole label set in one encoder pass on the owner's hardware, with single-label, multi-label, hierarchical and few-shot modes. It returns label scores with no calibration claim, the bundled server has no authentication, and the last three test runs on the main branch, on 24 September 2026, failed.

Drex 1.5

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.

Before you call either

GLiClass

  1. Pass --host 127.0.0.1 to python -m gliclass.serve, or put the port behind your own gateway. The server checks no credential
  2. On a machine without a GPU add --device cpu --dtype float32 --num-gpus-per-replica 0. The default configuration expects CUDA
  3. Send one text a request to POST /gliclass. An array in texts is cut to its first item without an error
  4. Set multi_label to false for one label from a set. The default scores each label independently, so scores do not sum to 1
  5. Keep text plus labels under the pipeline's 1,024-token max_length, or use ZeroShotClassificationWithChunkingPipeline. Longer input is truncated silently

Drex 1.5

  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.

Questions

Which is better for AI agents, GLiClass or Drex 1.5?

Drex 1.5 scores 69.7 (B) on agent readiness against GLiClass's 49.9 (D), and leads in 5 of 7 scored categories. GLiClass leads on payments & pricing.

Do GLiClass and Drex 1.5 need an API key?

GLiClass needs no key. Drex 1.5 needs an API key.

Can an agent call GLiClass and Drex 1.5 without installing anything?

No hosted endpoint is listed for GLiClass. Drex 1.5 has a hosted endpoint at https://console.nace.ai/v1/systemone.

Are GLiClass and Drex 1.5 open source?

GLiClass is open source (Apache-2.0 (library and the model weights we checked)). No open-source release is listed for Drex 1.5.

Other comparisons with GLiClass or Drex 1.5

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.