GA
Research lab · Product company
AI for business, built end to end
Frontier AI models, team apps, and developer APIs — on state-of-the-art GPUs and servers we own. One company, one stack.
Product portfolio
What we ship
Fifteen products across three layers — apps, APIs, and infrastructure. Start with one or run the full stack.
Simon picks the best model for you
api-call.ts
fetch("/v1/chat", {
method: "POST",
body: {
model:
input: "..."
})
method: "POST",
body: {
model:
input: "..."
})
fetch("https://api.geniuspro.io/v1/chat/completions", {
method: "POST",
headers: { "Authorization": "Bearer YOUR_KEY" },
body: JSON.stringify({
model:
messages: [{ role: "user", content: "..." }]
})
})
method: "POST",
headers: { "Authorization": "Bearer YOUR_KEY" },
body: JSON.stringify({
model:
messages: [{ role: "user", content: "..." }]
})
})
Every request is encrypted and private.
Apps & workspaces
Products teams use every day — chat, studio, your brand, and the platform control plane.
Preview
Beta
Beta
GA
API & models
Simon, our model lineup, modalities, and composable workflows for builders.
GA
GA
Models
v2026.06Our lineup of AI models — experts, orchestrators, and direct frontier access.
View productGA
GA
GA
GA
GA
Preview
Infrastructure
Agent fleet, private GPU compute, and our own cloud — infrastructure we operate.
Preview
Preview
Compute
PreviewPrivate GPU inference on lab hardware — dedicated nodes, not shared cloud.
View productHybrid stack
MicrosoftGoogleAmazonGroqNVIDIAIBMOracleCloudflareSupabaseVercelOpenAIAnthropic
US cloud API · lab GPUs · infrastructure we operate
Apps, API, and models — one company
Ainslie for your team. Simon for your product. Models trained and shipped from GeniusPro Lab — same stack, one bill.
GeniusProSimon Says Models
OpenAIGPT Models
AnthropicClaude Models
GoogleGemini Models
MetaLlama Models
MicrosoftPhi Models
NVIDIANemotron Models
AWS
BedrockNova Models
Arcee AITrinity Models
PerplexitySonar Models
You
Lab-maintained by default.Your choice when you need it.
We benchmark, fine-tune, and upgrade models behind stable slugs. Use Ainslie out of the box or drive every knob through Simon and Platform.
We maintain
Models that stay current
Simon routes each request
Classifies the prompt and picks the best model for the task — coding, reasoning, design, or agents — with automatic failover.
Auto-update slugs
Pin simon-says-chat or simon-says-coding once. We swap the model behind the slug after lab benchmarks pass — no migration projects.
Lab releases you can follow
New experts, orchestrators, and frontier access ship through GeniusPro Lab — often without changing your integration.
You control
Products and policies
Ainslie for teams
Chat, voice, studio, and models in one workspace — no API keys required for everyday employees.
Pin any model or build CATs
Call claude-opus-4-8, gpt-5.5, or gemini-3.1-pro directly — or publish cat:your-slug workflows from Platform.
Platform governance
Orgs, workspaces, per-client keys, spend caps, and audit logs — built for agencies and white-label resellers.
Three ways to call a model.
Automatic routing, stable task slugs, or direct frontier access — same Simon API, same Platform billing.
Automatic
Let Simon decide
model: "simon-says"
Send your prompt — Simon picks the best model for the task and fails over if a provider blips.
ChatCodingReasoningDesignAgentsVisionVoice
Default for most apps. No routing code to maintain.
Recommended
Recommended
Use task slugs
model: "simon-says-chat"
Pick a slug by job — chat, coding, finance, image. We keep it on the best version behind the name.
simon-says-chatsimon-says-websearchsimon-says-codingsimon-says-coding-balancedsimon-says-reasoningsimon-says-designsimon-says-agentssimon-says-financesimon-says-imagesimon-says-visionsimon-says-voice-standardsimon-says-tts
Auto-update slugs — upgrades happen in the lab, not in your repo.
Direct access
Pick any model
model: "claude-opus-4-8"
Call upstream models by real slug when you need a specific provider or reproducibility.
Anthropic9 models
claude-opus-4-8claude-opus-4-7claude-sonnet-4.6claude-haiku-4.5
OpenAI14 models
gpt-5.5gpt-5.4gpt-5.3-codexgpt-4.1
Google12 models
gemini-3.1-progemini-3-progemini-3-flash
xAI2 models
grok-3grok-3-mini
Meta2 models
llama-4-maverickllama-4-scout
Hundreds of slugs in the live catalog — browse models →
What teams build with GeniusPro
From employee workspaces to customer-facing apps — Ainslie, Simon, CATs, and Fleet on one stack.
Ainslie
Team AI workspace
Give every employee chat, voice, models, and studio in Ainslie — governed through Platform.
Simon API
AI-native products
Ship copilots and generative features with one OpenAI-compatible API — no multi-vendor SDK sprawl.
Auto-routing
Customer support bots
Route FAQs to budget slugs and escalations to premium models — failover is automatic.
White Label AI
Agency white label
Your brand, domain, and client portals — our models, billing, and US-hosted governance.
CATs
Multi-step pipelines
Chain models, vision, and image gen into cat:your-slug — edit from Platform, no redeploy.
Tiger + Fleet
Production agents
Long-running Tiger workflows that retry, checkpoint, and resume — plus Fleet workers you dispatch.
Platform
Built different — on purpose
Not a pass-through API. A lab that ships products: orchestration, workflows, modalities, and infrastructure we operate.
Simon orchestrator
One slug that picks the right model
Call "simon-says" — a router classifies your prompt, dispatches to the best specialist, and merges results when needed. One tier price. No routing logic in your app.
Auto-RoutingParallel DispatchCost OptimizationDirect Provider APIs
Auto-update models
Stable names. Better models over time.
simon-says-chat is a contract, not a frozen checkpoint. When the lab validates a better model, we swap it in — your integration keeps the same slug.
simon-says-chatsimon-says-codingsimon-says-reasoningsimon-says-design
CATs
Multi-step workflows. One API call.
Compose models, vision, and transforms in Platform — invoke as cat:your-slug from Simon or Ainslie. Edit live, no redeploy.
LLM StepsVision (SAM3)Image GenJS/Python Transforms
Tiger Mode
Durable runs that survive failures
Checkpoints after every step, controlled retries, resume where you left off — for long-running agent and pipeline work.
Step PersistenceControlled RetriesResume After FailureProgress Tracking
Full-stack AI
Voice, vision, memory, search — one endpoint
Realtime voice, computer vision, image gen, web search, and session memory through Simon — the same API Ainslie uses under the hood.
Computer VisionReal-Time VoiceSession MemoryWeb SearchImage Gen
Developers
Build on Simon
One OpenAI-compatible endpoint — chat, voice, vision, models, and composable CAT workflows.
Infrastructure
Our cloud, our GPUs — not just theirs
Our own US cloud for production API, plus GPU lab nodes for agents and inference. Hybrid by design — reliable infrastructure we operate, not a pass-through to someone else's datacenter.
Fleet
Dispatch agent workers on your machines or ours — monitor and scale from one hub.
View productCompute
Dedicated GPU nodes for private inference — lab hardware, not shared cloud.
View productWhat we do
Research, train, and ship models
Our lab is how Simon's lineup gets better — benchmarks, fine-tunes, and releases you can follow.
Visit the labGeniusPro Lab
Research, train, and ship models
Follow model releases, benchmarks, and product news from our research pipeline.
Models
New Simon Says experts: finance, legal, architecture
Call a domain slug — simon-says-finance, simon-says-lawyer, and more — without wiring your own prompts.
Jun 8, 2026
Read articleModelsAuto-update slugs: ship once, stay current
simon-says-chat and gp-auto-* pick up better models behind the same name. No migration projects.
Jun 5, 2026
Read articleProductWhite Label AI — your logo, our stack
Launch Ainslie-class experiences under your brand with workspaces, billing, and US hosting included.
Jun 2, 2026
Read articleFor business
Your data stays private
No training on your prompts. PII scrubbed in logs — and opt-in scrubbing before upstream. US providers. Built for teams, agencies, and white-label customers.
No Training On Your Data
Your prompts pass through to the model you chose and nothing is used to train anyone's models — not ours, not the upstream provider's.
PII Scrubbed From Our Logs
Emails, phone numbers, SSNs, credit cards, IP addresses and API keys are masked before requests land in our logs and stored history.
Opt-In Edge Scrubbing
Send X-GP-Scrub-Prompt: true (or scrub_pii: true in the body) and PII is masked before the request ever reaches the upstream provider.
US-Based AI Providers
Every model routes through a US-hosted company. Friendly to SOC 2, ISO 27001, and HIPAA-regulated workloads.
Always Encrypted
Requests are encrypted in transit over TLS and at rest in our managed infrastructure.
Fully Auditable
Every request logged. Usage, cost, and routing decisions tracked for your team.
Grows with your org
Solo builder, scaling startup, or multi-client agency — Platform gives you orgs, workspaces, and client views from one control plane.
Organizations
One roof, full control
Invite teams, manage billing, and see usage across Ainslie, Simon, and API keys — one relationship, not six vendors.
Workspaces
Isolate every team or client
Each workspace gets its own keys, balance, and logs. Bill per project, per product, or per customer.
Client portal
Give clients their own view
White-label resellers expose usage and spend per client — without sharing your root org or API keys.
See what you save
Usage-based pricing on Simon — compare direct provider spend vs routed simon-says-* at your monthly volume.
1B tokens
10M50M100M250M500M1B2.5B5B10B
Claude Opus 4.8
65%Direct
$10,000/mo
Simon
$3,463/mo
You save
$6,538
Claude Opus 4.6
65%Direct
$10,000/mo
Simon
$3,463/mo
You save
$6,538
GPT-5.5
69%Direct
$11,250/mo
Simon
$3,463/mo
You save
$7,788
GPT-5.4
38%Direct
$5,625/mo
Simon
$3,463/mo
You save
$2,162
Claude Sonnet 4.6
42%Direct
$6,000/mo
Simon
$3,463/mo
You save
$2,537
Gemini 3.1 Pro
23%Direct
$4,500/mo
Simon
$3,463/mo
You save
$1,037
| Model | Direct cost | Simon | You save |
|---|---|---|---|
| Claude Opus 4.8 | $10,000/mo | $3,463/mo | $6,53865% |
| Claude Opus 4.6 | $10,000/mo | $3,463/mo | $6,53865% |
| GPT-5.5 | $11,250/mo | $3,463/mo | $7,78869% |
| GPT-5.4 | $5,625/mo | $3,463/mo | $2,16238% |
| Claude Sonnet 4.6 | $6,000/mo | $3,463/mo | $2,53742% |
| Gemini 3.1 Pro | $4,500/mo | $3,463/mo | $1,03723% |
Assumes 75% input / 25% output. Blended simon-says rate across a typical workload mix. Per-slug rates on /developers and /models. Billed through Platform.
Get started
Ready when you are.Pick your path.
Ainslie for your team. Simon for your product. Talk to us for white label, Fleet, Compute, and lab previews.
For teams
Try Ainslie
Chat, voice, models, and studio — your team's AI workspace.
Open AinslieFor developers
Build on Simon
One OpenAI-compatible API — models, voice, vision, and CAT workflows.
Start buildingFor business
Talk to our team
White label, private GPUs, Fleet, Compute, and preview products.
Get in touch