Model tracker / Decision models / Nace.AI
Drex 2.0
Open-weight 35B mixture-of-experts decision model (about 3B active) on Qwen3.6-35B-A3B
Drex 2.0 is an open-weight decision model from Nace.AI (Nace.AI Open RAIL-M: organisations above $1M in revenue or funding need a separate commercial licence (personal and research use excepted), as do competitors of Nace; bars fully automated decisions with significant effects on people without human review, 35B total, about 3B active per token parameters): it answers typed questions, such as yes or no, pick one or pick several, with probabilities instead of writing text. It costs $0.06 per million input tokens (Nace Console API), and output is free. JevBench, the main independent benchmark, doesn't list it yet.
- Made by
- Nace.AI
- Weights
- Open, Nace.AI Open RAIL-M: organisations above $1M in revenue or funding need a separate commercial licence (personal and research use excepted), as do competitors of Nace; bars fully automated decisions with significant effects on people without human review
- Size
- 35B total, about 3B active per token
- Where
- Nace Console API (the default model and drex-latest), OpenRouter (served by DeepInfra)
- Images
- No, text only
- Context
- 131,072 state tokens hosted (262,144 with the longest question); self-hosted default 16,384, raisable to 131,072
- Fine-tuning
- Tuning on your labelled decisions on request from Nace, with the weights shipped to you; open weights
Drex 2.0 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 |
|---|---|---|---|---|
| Nace Console API (the default model and drex-latest), OpenRouter (served by DeepInfra) | $0.06 | Free | Output is free. The same price on OpenRouter. Released 6 October 2026. A Q8_0 GGUF (36.9 GB) runs with Nace's forks of llama.cpp and Ollama. | 2026-11-07 |
Drex 2.0 speed
| Measure | Result | Reported by |
|---|---|---|
| Reported speed | 151 ms median API time per decision, network included | Nace.AI |
Measured by the maker on its own hardware; JevBench doesn't list this model yet.
Decision models like Drex 2.0
Every other decision model we track, by independent JevBench score.
| Model | Input / M | JevBench | Weights |
|---|---|---|---|
| Drex 2.0 | $0.06 | – | Open |
| Mammoth-1 Baseten (Base Labs) | $0.10 | 85.2 | Open |
| Drex Max v1.1 Nace.AI | $0.03 | 84.2 | Closed |
| Perplexity Decider V1.2 Perplexity | $0.02 | 80.8 | Open |
| d1-ultra Liquid AI | $0.042 | 80.5 | Closed |
| Kev-27B Jared Palmer | Self-host | 78.9 | Open |
| Clef Cloudflare | $0.24 | 75.3 | Open |
| Jev 1.14 TypeSafe AI | $0.042 | 75.0 | Closed |
| d1 Liquid AI | $0.02 | 74.4 | Closed |
| Mercury Decide Inception | $0.02 | 73.6 | Closed |
| OpenAI Decisions (GPT-6 Luna) OpenAI | $0.10 | 73.5 | Closed |
| Microsoft-Decision-1 Microsoft | $0.042 | 70.8 | Closed |
| Elephant-1 Baseten (Base Labs) | $0.035 | 70.2 | Open |
| Strands Decider 26B-A4B AWS Strands Labs | Self-host | 66.7 | Open |
| Clef-flash Cloudflare | $0.038 | 63.8 | Open |
| Perplexity Decider Flash V1.2 Perplexity | $0.01 | 61.7 | Open |
| Kev-9B Jared Palmer | Self-host | 61.5 | Open |
| Clef-omni Cloudflare | $0.15 | 58.3 | Open |
| Joey-1 Baseten (Base Labs) | $0.01 | 57.8 | Open |
| d1-3B Liquid AI | Self-host | 49.5 | Open |
| d1-omni-600M (experimental) Liquid AI | Self-host | 32.8 | Open |
| Laya Convai Innovations | $0.011 | 32.5 | Open |
| CLM-8B Stanford and NVIDIA | Self-host | 20.1 | Open |
| Clairic Cloudflare | $0.10 | – | Closed |
| d1-2B Liquid AI | Self-host | – | Open |
| d1-omni-2B Liquid AI | Self-host | – | Open |
| Drex 1.5 Nace.AI | $0.05 | – | Open |
| Kev-0.8B Jared Palmer | Self-host | – | Open |
| Kev-4B Jared Palmer | Self-host | – | Open |
| Koa Decisions Salesforce | Self-host | – | Closed |
| Strands Decider 12B AWS Strands Labs | Self-host | – | Open |
| Strands Decider 2B AWS Strands Labs | Self-host | – | Open |
| Strands Decider E2B AWS Strands Labs | Self-host | – | Open |
| Strands Decider E4B AWS Strands Labs | Self-host | – | Open |
Compare them all in the decision models table →
Drex 2.0 on dbggr
- Stop asking your LLM multiple-choice questions: a guide to decision models Article, 2026-10-02
Drex 2.0 FAQ
How much does Drex 2.0 cost?
Drex 2.0 costs $0.06 per million input tokens at Nace Console API (the default model and drex-latest), OpenRouter (served by DeepInfra). Output is free. The same price on OpenRouter. Released 6 October 2026. A Q8_0 GGUF (36.9 GB) runs with Nace's forks of llama.cpp and Ollama. Checked 7 November 2026.
Is Drex 2.0 open weight?
Yes. Nace.AI publishes the weights under Nace.AI Open RAIL-M: organisations above $1M in revenue or funding need a separate commercial licence (personal and research use excepted), as do competitors of Nace; bars fully automated decisions with significant effects on people without human review, so you can run and fine-tune it yourself.
How fast is Drex 2.0?
Nace.AI reports 151 ms median API time per decision, network included, measured on its own hardware.
How does Drex 2.0 compare with Jev?
JevBench doesn't list Drex 2.0 yet, so there's no independent comparison. Jev costs $0.042 per million input tokens; Drex 2.0 costs $0.06.