1. Ward
  2. /
  3. Product
  4. /
  5. LLM Agnostic
Product · LLM Agnostic

Any LLM you choose,
routed per task and audited per query.

Today’s leading model may not lead next quarter. Ward runs on OpenAI, Anthropic, Gemini, or your own, routed automatically and logged on every call.

Get a demo →

The right model for every task

Not every question needs the most expensive model. Ward’s router sends each query to the right LLM: fast models for lookups, reasoning models for root cause. Forecasting numbers come from classical models, not the LLM. The LLM frames the answer and cites the math.

app.getward.ai Live demo
Acme Retail @Merchandising: VP Analyst claude-sonnet default
A

Chat

Ask anything. Ward routes to the right agent and returns cited answers.

Why did Store 37 miss target last week?
You · 9:42 AM
Schema Scout · routed to Merchandising Agent

I pulled Store 37’s last 28 days against the chain baseline. Two root causes, both compounding.

SignalFinding
labor_efficiencyRev/labor-hour −22% vs. cluster, staffing mismatch at 11a–1p peak
inventory.freshFresh fill 83%, backroom replenishment lag at 2–4p
promo.liftBOGO crackers cannibalized Brand Y by 28%, net category +6%

Recommend: re-baseline Store 37 schedule against true peak, raise replen window to 1p, and review the BOGO before next cycle.

8 parallel queries 3 sources cited confidence 0.92
Show me how to fix the staffing mismatch.
You · 9:43 AM
Labor Agent · drafting schedule diff
Querying labor_scheduling…
› Ask anything, Ward routes to the right agent. Cmd+K

AI Insights

Agents run against your baselines overnight. These are what they flagged without being asked.

3 flagged 8 queries run 3 sources swept 04:00, acme-retail
Act now labor_efficiency 0.94

Rev/labor-hour −22% vs. cluster, staffing mismatch at 11a–1p peak

Schema Scout · routed to Merchandising Agent Pin Ask Ward Investigate
Review inventory.fresh 0.89

Fresh fill 83%, backroom replenishment lag at 2–4p

Promo Agent · routed via Merchandising Pin Ask Ward Investigate
Watch promo.lift 0.81

BOGO crackers cannibalized Brand Y by 28%, net category +6%

Margin Agent · routed via Finance Pin Ask Ward Investigate
Recommended

Re-baseline Store 37 schedule against true peak, raise replen window to 1p, and review the BOGO before next cycle.

Reporting

Pinned views built from saved data-lake queries. Every number re-derivable from its SQL.

7d13w52w
Revenue vs. forecast
$48.2M
+4.2% WoW
Gross margin %
24.1%
−3.2pp
Fill rate, fresh
83.4%
−4.1pp
Shrink, West region
2.41%
+0.8pp
Revenue vs. forecast 13 weeks actual, 6 weeks forecast, 80% interval
Actual Forecast 80% interval
% of plan 106 94 100 forecast → W−13 W−5 today +6wk
Holt-Winters + weather regressor MAPE 4.1% at 4wk Backtested 24 months Crosses plan in 3 weeks
Forecast error by horizon MAPE, 24-month backtest
1wk 2.1%
2wk 3.0%
4wk 4.1%
8wk 6.3%
13wk 8.9%
Accuracy bar for promo decisions: ≤5% at 4wk
Models in production Every forecast ships a model card
ModelHorizonMAPE
holt_winters4wk4.1%
arima_sarimax13wk8.9%
gbm_demand1wk2.1%
bayes_hiernew store11.4%

Models

Browse, search, and manage data–lake model definitions for your tenant.

Search models…
NameNamespaceVersion
sap_pos_transactionssap1.0
sap_inventory_shrinkagesap1.2
sap_labor_schedulingsap1.0
retail_inventory_weeklyretail1.1
retail_google_ads_dailyretail1.0
retail_meta_ads_dailyretail1.0
retail_ga4_website_dailyretail1.0

Sources

Connect external systems to the data lake.

Search sources…
NameTypeLast sync
sap_pos_transactionsimport2m ago
sap_inventory_shrinkageimport2m ago
sap_labor_schedulingimport14m ago
retail_inventory_weeklyimport1h ago
retail_google_ads_dailyimport1h ago
retail_meta_ads_dailyimport1h ago
retail_ga4_website_dailyimport1h ago

Architecture

Two ways to connect. Federate against your live systems, or ingest into Ward’s data lake. Toggle below.

Your systems · read-only
SAP Retail
Snowflake
BigQuery
Shopify
NCR Voyix
Ward Gateway
TLS 1.3 · AES-256
Querying live · data stays put
Federated answers
SELECT * FROM sap.pos
JOIN retail.inventory_weekly
WHERE store_id = 37
→ insight cards
Ward Data Lake
→ baselined per store
TLS 1.3 in transit AES-256 at rest Read-only credentials SOC 2 Type II underway VPC peering · PrivateLink

Pipelines

Move data from sources into models on a schedule.

NameSourceModelStatusSchedule
sync_sap_pos_transactionssap_pos_transactionspos_transactionsenabledhourly
sync_sap_inventory_shrinkagesap_inventory_shrinkageinventory_shrinkageenableddaily
sync_sap_labor_schedulingsap_labor_schedulinglabor_schedulingenableddaily
sync_retail_inventory_weeklyretail_inventory_weeklyinventory_weeklyenabledweekly
sync_retail_google_ads_dailyretail_google_ads_dailygoogle_ads_dailyenableddaily
sync_retail_meta_ads_dailyretail_meta_ads_dailymeta_ads_dailyenableddaily

Streams

Real-time ingestion pipelines.

0events / min
0streams active
0% delivered
  • pos.txn store_037, basket $42.18
  • inv.move dc_west → store_104
  • labor.clock store_022 shift_start
  • pos.txn store_211, basket $19.04

Policies

Browse and manage Cedar access policies for your tenant.

TLS 1.3 AES-256 Read-only SOC 2 II
Policy IDEffectResources
merch-read-defaultpermitModel::*
finance-read-shrinkagepermitModel::"inventory_shrinkage"
vendor-blockedforbidModel::"labor_*"
region-west-onlypermitTenant::"acme"

Entities

Principals and resources referenced by Cedar policies.

Search UIDs…
Entity UIDTypeTenant
Tenant::"acme-retail"Tenantacme-retail
Model::"sap.pos_transactions"Modelacme-retail
Model::"sap.inventory_shrinkage"Modelacme-retail
Model::"sap.labor_scheduling"Modelacme-retail
Model::"retail.inventory_weekly"Modelacme-retail
Model::"retail.google_ads_daily"Modelacme-retail

Providers

Manage LLM API keys and the model profiles that use them.

API Keys Model Profiles
NameProviderUsed byCreated
anthropic-defaultAnthropic3 profilesApr 22
openai-defaultOpenAI2 profilesApr 22
gemini-defaultGemini1 profileApr 22
ollama-onpremOllama2 profilesApr 22

LLM-agnostic. Bring your own key, route per task. No lock-in.

Settings

Manage your dashboard preferences and account.

Appearance
Theme • Light ° Dark

Light and dark themes are available. Your choice is remembered per browser.

Account
NameAdmin
Emailadmin@acme.io
Tenantacme-retail
LLM-agnostic. Anthropic, OpenAI, Gemini, Ollama. Bring your own keys.
Multi-LLM, routed per task
Routes to OpenAI, Anthropic, Gemini, or open-source. No lock-in, switch any time. Fast models for lookups, reasoning models for root cause, specialized math for forecasts. Every route logged with cost and latency.
OpenAI · Anthropic · Gemini · Open-source · Cost-efficient
Bring your own LLM
Already have an enterprise AI agreement? Bring your API keys. Ward runs on your models, in your environment. You hold cost, compliance, and data residency.
BYOLLM · Your API keys · Your data

Audit any query, any answer

The reasoning graph shows every step: which tables were queried, what was computed, which forecast model ran, which LLM answered. Click any number to walk it back to the row. Hand the trail to compliance, no matter which model powered the analysis.

Charters on file. Changes on the record.

Every agent ships with a written charter: scope, sources, allowed actions, owner, version. Diff two versions side by side. Every edit is logged with name, time, ticket, and approver. Roll back in one click. Export to your SIEM.

Cyber liability covered

Cyber and tech E&O policy on file, AI rider included. Certificate of insurance to procurement in a day. Coverage names data breach, regulatory response, and AI-specific liability.

One platform, every LLM, zero lock-in. No single model stays on top for long, so Ward’s router means you don’t have to pick a winner. The math runs on classical models, the LLM frames the answer, and the audit trail covers both.
WHO
Roles that depend on it

Built for the teams
that catch problems first.

Director / VP
Head of IT
Technology
VS
See it. Or compare it.

Operator stories
and the alternatives.

PLAY
Plays in production
How operators run llm-agnostic on Ward today.
DBX
Ward vs Databricks
Why building this on top of your lakehouse stalls.

Two ways to start.

Run a fixed-fee pilot on your data, or talk to advisory about a broader engagement.

Get a demo → Talk to advisory
Spend governance

Control AI spend by department and user.
Without skimping on the outcome.

Give every department and every user a compute budget. Ward routes each question to the cheapest model that clears the quality bar, so finance caps the spend without capping what the business gets back.

Compute budgets, this month 68% used · on pace
Merchandising
14,200 / 20,000
Supply Chain
9,800 / 15,000
Store Ops
6,100 / 10,000
Ecommerce
4,700 / 5,000
Finance
3,400 / 5,000
Per-user caps. Alerts at 80%. Hard stop or overage approval, your call. Ecommerce flagged at 94%, before it became a surprise invoice.

Your AI vendor shouldn’t lock in your model.

Ward works with any LLM. Switch anytime. See it on your data.

Get a demo → Take the 3-minute assessment

Read-only to start · your LLM keys · SOC 2 Type II underway · or book a call directly

Get started

Find out what your data has been hiding.

Tell us about your operation. We’ll show you the problems Ward catches, and the ones your current tools miss.

Step 1 of 3
What are your goals?
Step 2 of 3
About your operation
Step 3 of 3
Your contact info