The execution layer for agents.
Your agent writes the code. Nace does the grunt work around it: reading files, sorting, and making the small calls. It runs on fast, cheap models.
Works with
Leading performance at a fraction of the cost.
| Model | Creator | Type | Decision Index | Cost per 1M Tasks (USD) |
|---|---|---|---|---|
| Claude Haiku 5.5 (Max) | Anthropic | Language model | 76.5 | $6.2k |
| GPT-6 Luna (Max) | OpenAI | Language model | 75.7 | $1.5k |
| Drex Max v1.1 | Nace.AI | Decision model | 72.6 | $103 |
| SubQ Snap 1.2 | Subquadratic | Decision model | 72.6 | $137 |
| GLiDE v1.2 | Fastino | Decision model | 72.5 | $780 |
| Mammoth 27B | Base Labs | Decision model | 71.4 | $264 |
| d1-ultra | Liquid AI | Decision model | 70.6 | $133 |
| Decider v1.3 | Perplexity | Decision model | 70.4 | $68 |
| GPT-6 Luna (Low) | OpenAI | Language model | 68.8 | $485 |
| Claude Haiku 5.5 (Low) | Anthropic | Language model | 67.2 | $861 |
| d1.1 | Liquid AI | Decision model | 66.5 | $63 |
| Elephant 9B | Base Labs | Decision model | 66.4 | $92 |
| Jev 1.14 | TypeSafe | Decision model | 66.2 | $156 |
| Gemini Decision 1 | Decision model | 66.0 | $345 | |
| Decider Flash v1.3 | Perplexity | Decision model | 64.1 | $34 |
| Microsoft-Decision-1 | Microsoft | Decision model | 62.1 | $143 |
| Mercury Decide | Inception | Decision model | 62.1 | $70 |
| Celeris-1.1 Decision | Celeris | Decision model | 62.0 | $14 |
| Drex 2.0 | Nace.AI | Decision model | 61.6 | $200 |
| Clef | Cloudflare | Decision model | 60.3 | $863 |
| GPT-6 Luna (Decisions) | OpenAI | Decision model | 60.2 | $339 |
| Clef-Omni | Cloudflare | Decision model | 57.8 | $529 |
The frontier joins the models no other model beats on both score and cost. Hover a point for its numbers.
Builders on Drex and Document Intelligence, in their words.
- VVishesh Baghel@VisheshBaghell · Sep 25
$499,235 of unbilled work across 1000 work orders. @NaceAI's drex found $482,260.
drex reads the note and the invoice and flags what's missing. code does the math. when drex isn't sure, a human decides.
- AAakash Kanojiya@Kanojiyaaakash1 · Oct 8
I threw a messy PDF at Nace's new Document Intelligence model.
Instead of just converting the PDF into plain text, NDI is built to understand the document's structure, extract useful information, and connect every extracted value back to its original source.
It's a perception layer built specifically for AI agents.
- IIsrafil@Israfilv2 · Sep 30
Nace shows Drex playing a strategy game. i gave it a harder opponent: my timeline
Spoiler Guard sends each post to the new decision model and blurs the spoilers, 352 post checked in one session with sub-second calls
- DDean W. Perkins@deanwperkins · Oct 6
someone wired a small open model to a real security camera
389 ms per call
the vision model sees, drex decides
- IIvan Fioravanti@ivanfioravanti · Sep 25
Drex: faster, cheaper, better than jev, 32k context and with open weights coming soon? I will surely give it a try!
- SSuperSaiyanColtSep 26
I’m using Drex as a fast decision layer inside my project Elyreon/NOVA mainly for routing tasks, choosing tools/models, classifying failures, retries, urgency, and handoffs.
the bot already responds sub 1s so i cant imagine how much better drex will make it
- EEduardo OrdaxSep 25
Using a massive LLM to decide which tool to call feels like hiring a CEO to sort your inbox.
Drex targets those decisions: routing, ranking, classification and guardrails.
- AahmadghoOct 7
i was using jev for playing vs a bot in an old game called Little fighter 2
drex 1.5 is actually very comparable with jev's behaviour in game infact might be better
also it's less latent and very fast i've dropped jev eversince and been optimizing for drex
- NNawi@Aria_Nawi · Oct 8
Earnings day is chaos. One filing, a mountain of invoices and a feed of X posts. One agent read all of it
@NaceAI Document Intelligence read all three, and every value it pulls links back to where it sits
the part I'd watch: when there isn't enough data, Drex says "insufficient data" instead of guessing
- KKai@hqmank · Sep 25
Switching was easy. I pointed my existing Jev setup at Drex, kept the same SDK, changed 3 environment variables, and it worked.
I see a good fit for agent routing, tool selection, document classification, and choosing the next action in browser automation.
- MMAX@MAXdeg0 · Oct 8
i built a full NDI 1.0 trading system and open-sourced the entire repo.
NDI extracts the information, grounds it back to the original source, then evaluates the decision while a frontier model reasons over the context.
market data + financial documents + AI reasoning → trading decision.
- GGipp@gippp69 · Sep 30
128k context means it can route with real state instead of tiny summaries.
the interesting part isn’t more generation. it’s better decisions before generation starts.

Watch an agent run the workflow.
An agent with a Nace key reads an AP packet, finds the page that matters and decides what to do with it. Every call in the window is a Nace endpoint, and every value comes back with the page it was read from.

$ export NACE_API_KEY="nace_sk_…"
$ claude
> This AP packet from Halden Supply mixes invoices, a contract amendment and a W-9. Which invoice can we pay today, and did the terms change?
console · split(file: "halden-packet.pdf")
38 pages → 5 documents
1 Invoice INV-2291 p. 1–3
2 Invoice INV-2304 p. 4–6
3 Contract amendment No. 3 p. 7–19
4 W-9 p. 20
5 Purchase order PO-77120 p. 21–38
console · classify(document: 3, classes: "procurement")
contract_amendment · 0.97
console · extract(document: 3, schema: { payment_terms, effective_date, parties })
payment_terms "Net 45, 2% discount within 10 days" found
effective_date "2026-10-01" found
parties Halden Supply Co., Meridian Logistics found
console · ground(document: 3, text: "Net 45")
page 11 · box 0.12 0.41 0.88 0.47 · exact · 0.99
console · extract(documents: [1, 2], schema: { invoice_no, total, payment_terms, po_number })
INV-2291 $18,420.00 "Net 30" PO-77120 p. 2
INV-2304 $9,870.00 "Net 45" PO-77120 p. 5
console · systemone(state: "INV-2304 · $9,870.00 · Net 45 · matches PO-77120 · terms per amendment No. 3", questions: { approve: noul, route: choice })
approve noul 0.93
route payments 0.88 · review 0.09 · reject 0.03
Pay INV-2304 today: $9,870.00 against PO-77120 on Net 45, the terms Contract amendment No. 3 sets on page 11, effective 1 October 2026. Drex puts 93% on approving it and routes it to payments.
Hold INV-2291. It still carries Net 30 on page 2, which predates the amendment. Want me to flag it for AP review?
Why now
After the code, the run.
Code generation was the last wave. Workflow execution is the next. Once the code exists, it has to run: call tools, read files, classify, decide and reason, thousands of times a day. Paying a frontier LLM for every one of those steps is slow and expensive. That is where Nace sits.
The last wave
Code generation
Large models write the code, the tests and the pull request.
The next waveNace runs here
Workflow execution
The code calls tools, reads files and makes decisions, step after step, in production.
Three layers, one key
Perception, Decision, Search.
Three layers your agent calls instead of a frontier LLM. Each one is a small model built for one job, and all three sit behind the same key.
01 · PerceptionExample Document Intelligence extraction with its source Invoice 2023T210
Bookkeeping 600.00
Total 600.00/totalp. 1 · box 0.62 0.81 0.94 0.85
NDI
Every answer brings receipts.
Reads 36 file types, from scans to spreadsheets, email and video. Parse, split, classify, extract and ground, and every value points back to the page it came from.
02 · DecisionExample Drex decision: one invoice, three options INV-2304 · $9,870.00 · matches PO-77120
- approve0.93
- review0.05
- reject0.02
one pass
Drex
Odds, not opinions.
A calibrated probability on every option in one pass, with a median API time under 200 ms. Drex Max decides from screens, images and video.
03 · SearchExample Drex for Search query over a workspace - halden-amendment-3.pdf
- halden-msa.docx
- po-77120.xlsx
What are Halden's payment terms now?
Net 45, 2% discount within 10 days
halden-amendment-3.pdf · p. 11
Drex for Search
Drex finds the page.
Point it at where your files already live. Drex classifies the query, squeezes the search space to the few places the answer can live, and your agent gets the exact answer with the source page attached.
Frontier · Drex Max
Reads the screen and picks the next click, fast.
Drex Max is the frontier of the Drex family. Give it a screenshot and the question, and it puts a probability on every element your agent could click. Built for computer use and browser use, and it understands video too.
- Computer use
- Browser use
- Video understanding

- Invoice
- INV-88214
- Amount
- $16,920.00
- PO match
- 1,200 units
- Computer use
- 0.787On OSWorld-Verified, the benchmark of real tasks across desktop apps.
- Browser use
- 89.2On Mind2Web, the benchmark for picking the right element on a page.
- Every step
- <200 msMedian API time per decision, from text or a screenshot. Fast enough for every step of the loop.
One key, everywhere.
Console is plain HTTPS and JSON behind one nace_sk_ key. Any harness that can call a tool can call it; these are the ones we test with.
- 01
Create a key
Sign up in Console. Eligible new accounts get $5 of free credit.
- 02
Add Nace to your agent
One command gives your coding agent every layer as a tool.
claude mcp add nace -- uvx nace-mcp - 03
Call any layer
One header, one credit balance and one error envelope for Perception, Decision and Search.
Authorization: Bearer $NACE_API_KEY
Works where your agent works
- CodexCodex calls the same endpoints from its sandbox; the key travels as one environment variable.
- CursorWire Console into Cursor's agent so a document question is answered inside the editor.
- Claude CodeGive Claude Code a Console key and it splits, extracts and decides from the terminal.
- pipi's minimal harness fits Console's minimal contract: JSON in, probabilities and locations out.
- HermesHermes agents read files and route decisions through Console without a second integration.
- opencodeopencode reaches Console over plain HTTPS; no plugin to install, nothing to keep in sync.
One platform. Four boundaries.
Same models, same API at every tier. Only the boundary around your data moves.
01
Managed Cloud
Our GPUs, our API, a key from Console.
- Runs in
- Our cloud
- Network
- Public TLS
- GPUs
- Shared
- Updates
- Continuous
02
Dedicated Cloud
Your own cluster in our cloud, reached over private links.
- Runs in
- Our cloud
- Network
- Private link
- GPUs
- Dedicated
- Updates
- Continuous
03
Your Cloud
Installs in your AWS or Azure account. Data stays inside.
- Runs in
- Your cloud account
- Network
- Inside your VPC
- GPUs
- Dedicated
- Updates
- From registry
04
On-Prem & Air-Gapped
Your datacenter, your hardware, zero outbound.
- Runs in
- Your datacenter
- Network
- None
- GPUs
- Dedicated
- Updates
- Offline bundles
Questions
It's the set of models your agent calls while running a workflow, instead of sending every step to a frontier LLM. Document Intelligence reads files, Drex makes decisions, and Drex for Search finds the page in your files, all behind one API key.
Drex for Search finds the exact answer in your files and returns it with the file and page it came from. You upload a folder into a workspace, and Drex narrows each search to the places the answer can live. It works through the CLI, SDKs, MCP and REST.
Pricing is usage-based, from one credit balance. Eligible new accounts get $5 of credit, valid for 3 months, and every call, whether a decision or a document job, draws from the same balance. The price sheet is in the docs.
Document Intelligence reads 36 file types, from PDFs and scans to spreadsheets, email, audio and video. Parse turns any of them into Markdown, text or layout blocks, and every extracted value comes back with the page and region it was read from.
No, at any tier. Your data trains your own models only when you ask, and the weights, adapters and eval sets stay yours.
Yes. nace-mcp works with Claude Code, Codex, Cursor and other MCP clients, nace-cli works from the shell, and nace-sdk covers Python and TypeScript. Underneath it is plain HTTPS and JSON, so any harness that can call an API can call Nace.
Yes. Choose Dedicated Cloud over a private link, your own AWS or Azure account, or air-gapped in your data center. The models, API and console are the same at every tier.
Talk to salesFirst request today. Your own cloud when you need it.
Create a key in Console and send your first request in a minute. When the data cannot leave your network, we bring the platform inside it.
Are you an agent? Read this page as Markdown