Supercharge your agentic search with Decision‑powered retrieval.
Your coding agent gets 6% more answers right with 62% fewer steps and 56% fewer tokens, and finishes 2.3× faster.
OfficeQA Pro
DefaultWith Drex for Search
| Harness | Model | Effort | Accuracy as shipped, % | Accuracy with Drex for Search, % | Gain, points | Agent turns as shipped | Agent turns with Drex for Search |
|---|---|---|---|---|---|---|---|
| Gemini CLI | Gemini 3.1 Pro | medium | 54.9 | 61.7 | 6.8 | 22.9 | 10.4 |
| Hermes Agent | Qwen3.8-27B | medium | 56.4 | 60.2 | 3.8 | 20.1 | 6.5 |
| pi | Opus 5.5 | medium | 65.4 | 68.4 | 3.0 | 7.7 | 3.7 |
| Codex | GPT-6 sol | max | 66.2 | 70.7 | 4.5 | 14.3 | 8.6 |
| Hermes Agent | Opus 5.5 | medium | 64.3 | 68.4 | 4.1 | 8.5 | 3.7 |
| OpenCode | Qwen3.8-27B | medium | 51.1 | 53.4 | 2.3 | 23.7 | 10.0 |
| pi | Qwen3.8-27B | medium | 43.6 | 45.9 | 2.3 | 24.3 | 11.1 |
| Claude Code | Opus 5.5 | max | 66.5 | 69.9 | 3.4 | 18.5 | 9.1 |
| Codex | GPT-6 luna | max | 54.9 | 53.4 | -1.5 | 16.8 | 11.3 |
Trusted by
- Big Fouraccounting firm
Search that knows where to look.
Most of a search's quality is decided before the search runs. Drex makes those decisions first.
Fewer places to look. The right ones.
- 01
Classify. Drex reads the question and decides, before any search runs, whether it needs a quick lookup or an agent that reads documents and queries tables.
- 02
Squeeze. Drex walks the workspace level by level: folder, file, section, page, and inside a sheet, table and column. Each level keeps only what can hold the answer. Depth is fixed, so a million files take the same number of steps as a thousand.
- 03
Answer. Only the pages and tables Drex kept reach the LLM. The answer comes back with the file and page.
Beyond RAG.
Every generation of search narrowed where the machine looks. Exact match found the string, keywords the term, BM25 the likely passage, embeddings the similar one. Drex decides where the answer can be before the agent reads a page.
- Exact matchthe string
- Keywordsthe term
- BM25scored keywords, Elasticsearch
- Semanticembeddings and RAG, similar meaning
- DrexClassifies the question in one sub-second decision.Narrows folder, file, section, page, then table and column.Fixed depth, so time does not grow with the number of files.
Illustrative counts. Measured results are in the benchmarks below.
Ingest once. Search forever.
Ingest. Documents, tables, media, JSON. Every page is OCR'd and laid out, and every sheet is kept as a table.
Link. A knowledge graph joins what the files say. A term in one file reaches its definition in another.
Search. Drex classifies the question, narrows the candidates, and the answer cites file and page.
| Path | Status | ||
|---|---|---|---|
| contracts/halden-amendment-3.pdfdocument · 11 pages | document | 11 pages | ingesting |
| reports/fy25-10k.pdfdocument · 212 pages | document | 212 pages | ingested |
| ledgers/tb-q3.xlsxtable · 4 sheets | table | 4 sheets | ingested |
| calls/earnings-q3.mp4media · 41 min | media | 41 min | ingested |
| crm/accounts.jsonjson · 1,208 records | json | 1,208 records | ingested |
| contracts/halden-msa.docxdocument · 38 pages | document | 38 pages | ingested |
- pages
- 2,391
- graph
- 1,204 nodes · 3,870 edges
- index
- 6 files
One workspace per project. A search never leaves it.
Isolated. Its own index and graph, built only from its own files. Delete the workspace and everything built from it goes.
Shareable. Invite members. Everyone searches and asks over the same files.
Cited. Every answer names the file and the page, row or minute it came from.
Project Birch4 members
folders
- data-room/38
- financials/6
- legal/9
files
- cim.pdf
- qoe-report.pdf
- cap-table.xlsx
- mgmt-call-2.mp4
isolated · 5,804 pages
Northwind Foods1 member
folders
- pos/14
- rate-cards/2
- vendors/5
files
- po-77120.xlsx
- po-77121.xlsx
- rate-card-2025.pdf
- vendor-msa-arla.pdf
isolated · 640 pages
Halden Group2 members
folders
- contracts/6
- redlines/9
- correspondence/14
files
- msa-v4-redline.docx
- sow-3.pdf
- rate-card-2026.xlsx
- counsel-notes.md
isolated · 1,120 pages
Halden Group3 members
folders
- contracts/2
- ledgers/1
- reports/1
- calls/1
files
- halden-msa.docx
- halden-amendment-3.pdf
- tb-q3.xlsx
- earnings-q3.mp4
isolated · 2,391 pages
Search that is built for knowledge work.
Four public leaderboards for search over documents. Ranks as published, including the ones we do not lead.
MADQA
Accuracy, LLM judge · 500 test questions · Higher is better
- Drex for Search: 94.7
- Progress: 94.3
- Mixedbread: 93.4
- Perplexity: 92.1
- ARRAY Spectrum: 91.8
- Distyl Button: 91.7
- BM25 search agent: 82.2
- Google File Search: 78.6
- OpenAI File Search: 49.6
- ColBERTv2 RAG: 40.2
Leaderboards read October 20, 2026.
Built for the files that break other search.
PDFs and scans
Scanned statements, filings, contracts. OCR and layout keep the tables inside them as tables. The answer points at the page.
Spreadsheets and tables
xlsx, csv, Parquet. General ledgers and journal entries stay whole, however long. The answer points at the row and cell.
Audio and video
Earnings calls, recorded meetings, mp3 and mp4. The answer points at the minute.
Office and text
docx, pptx, md, json, txt. One question runs over documents and sheets together.
Full list of supported files in the docs.
Your files stay yours.
You upload what you want searched. Nothing else leaves your storage.
No training. Customer content never trains, tunes or evaluates shared models.
Retention. Files stay in the workspace until you delete the workspace or your account.
Encryption. TLS in transit, encrypted at rest, stored in the United States.
SOC 2 Type II. Audited. Request the report
- Members
- a.serikova
- d.kim
- +1
3 members · search and ask
- Retention
Until deleted
- Training
Customer content is not used for training
- Compliance
SOC 2 Type II
- Encryption
TLS in transit · AES at rest
Runs where your files are allowed to be.
Same models, same API at every tier. Only the boundary around your workspaces moves. SOC 2 Type II audited at every one.
Request SOC 2
- Runs in
- Our cloud
- Network
- Public TLS
- GPUs
- Shared
- Updates
- Continuous
01
Managed Cloud
Our GPUs, our API, a key from Console.
- Runs in
- Our cloud
- Network
- Private link
- GPUs
- Dedicated
- Updates
- Continuous
02
Dedicated Cloud
Your own cluster in our cloud, reached over private links.
- Runs in
- Your cloud account
- Network
- Inside your VPC
- GPUs
- Dedicated
- Updates
- From registry
- Runs in
- Your datacenter
- Network
- None
- GPUs
- Dedicated
- Updates
- Offline bundles
Four ways in. One key.
Upload a folder and search it straight from your terminal.
Plug search into Claude Code, Codex, Cursor or OpenCode, and your agent uses it on its own.
Call search from any language or tool over the web.
One Console key for everything.
$ pipx install nace-cli$ nace login$ nace workspace-create fy25-audit$ nace upload ./fy25 --workspace <id> --ingest$ nace search "Total liabilities at year end?"Total liabilities at year end were $48.2M.evidences (1): reports/fy25-10k.pdf, page 41 "Total liabilities 48,214"
Frequently asked questions
What is Drex for Search?
Drex for Search finds the exact answer in your files and cites the file and page it came from. You upload documents to a workspace, and Drex narrows each search to the few places the answer can be. It works through the CLI, SDKs, MCP and REST.
How does Drex for Search improve BM25 and RAG?
Drex for Search tells BM25 and RAG where to look, so the LLM reads less and misses less. Drex reads the question, picks a quick lookup or an agent that reads documents and tables, and narrows the candidates to the few that can hold the answer.
Do my files have to leave my storage to use Drex for Search?
Only the files you choose to search. You upload a folder from a laptop, a synced bucket or a drive export to an isolated workspace, and the files stay there until you delete the workspace or your account. If your data can't leave your network, Drex for Search can run in your own cloud or data center.
What file types can Drex for Search handle?
PDFs and scanned documents, spreadsheets, Parquet, JSON, audio and video. Spreadsheets and Parquet files are kept as tables, so an answer can point to the exact row or cell.
Does Drex for Search work with my existing coding agent?
Yes. Claude Code, Codex, Cursor and OpenCode can call it as a tool through the CLI or the MCP server. On OfficeQA Pro, the ten agent setups we tested used 57% fewer input tokens with it, and eight of ten answered more questions correctly.
Does Drex for Search use my content to train models?
No. Customer content never trains, tunes or evaluates shared models.
Point Drex at your files.
Create a workspace. Upload a folder. Ask.
