Riven Chat
The authenticated reasoning workspace. Deep thinking, high-context research, multi-step analysis, and agent-driven work — for power users, teams, and enterprise.
One platform family for AI workspaces, browser integration, code execution, media generation, and intelligence analysis.
Five products. One unified gateway. Deployable on our cloud, your infrastructure, or fully managed on your behalf.
Try it now
Product family
Each product has a distinct purpose. None of them are the same app in a different skin.
The authenticated reasoning workspace. Deep thinking, high-context research, multi-step analysis, and agent-driven work — for power users, teams, and enterprise.
The browsing and distribution surface. Fast, immediate, exploratory. Built for free-tier access, wide reach, and AI-native browsing without friction.
The execution workspace for builders. Write code, automate workflows, deploy, and operate — with agentic capabilities built into every step.
The developer and integration gateway. Build on the Riven platform with a clean, OpenAI-compatible API surface, keys, usage controls, and full routing support.
The media generation studio. AI-native video creation, avatar workflows, and production-grade media output — paired naturally with Chat and API.
Model infrastructure
RivenAI runs a tiered model infrastructure — local GPU inference, warm model cache, on-demand access, and automatic multi-cloud fallback. All unified behind a single OpenAI-compatible endpoint.
Tier 1
Your hardware, zero latency. Models run directly on your infrastructure with full data sovereignty.
Tier 2
Pre-loaded models ready instantly. No cold starts, no waiting for model loading.
Tier 3
Any model, any size. Request what you need, when you need it — billed by usage.
Tier 4
18+ cloud providers as automatic fallback. If one fails, traffic routes instantly.
Models available
Cloud providers
API endpoint
Vendor lock-in
Use cases
Use Riven Chat to work through complex documents, compare sources, build structured arguments, and produce research outputs that require sustained reasoning over long contexts.
Relay brings AI into the browser natively. Search, retrieve, summarise, and interact with web content without switching context — accessible to anyone, no heavy setup required.
Write code, debug, refactor, and deploy with Riven Forge. Agents handle repetitive execution work while you stay focused on architecture and output — all in a GitLab-native workflow.
Stratoveil provides the structured workspace for analysts. Multi-panel views, geospatial context, and a mission-control interface designed for information-dense, high-stakes work.
Riven Studio handles the media generation layer — video creation, avatar workflows, and production-grade output. Use it standalone or invoke it programmatically through Riven API.
Riven API gives developers an OpenAI-compatible gateway to the full Riven platform. Add AI capabilities to your own products with consistent routing, usage controls, and key management.
Deploy Riven across your organisation with enterprise-grade identity, controlled access, audit logging, and platform services that scale from a small team to a full enterprise environment.
A multi-panel intelligence and exploration platform for analysts, researchers, and operators working with complex, layered information. OSINT workflows, geospatial analysis, and source intelligence — built for serious analytical work.
Stratoveil operates as a distinct brand under RivenAI. Same platform foundation, different market position — built for intelligence professionals who need a different kind of workspace.
Deployment
RivenAI runs wherever you need it — our cloud, your cloud, your hardware, or fully managed on your behalf. Same platform, same capabilities, your choice of control.
Hosted and managed by RivenAI. Start immediately with no infrastructure setup. Best for teams who want to move fast without ops overhead.
Get startedDeploy the full Riven stack on your own infrastructure — your VMs, your GPUs, your data center. Full control, full sovereignty.
Contact salesBig Sky operates Riven on your behalf — on-site or in your cloud. Enterprise SLAs, dedicated support, custom deployment.
Talk to usAccess
Add AI to your browser. Relay installs as a browser extension and brings the Riven platform into every page you visit. Free-tier access, no account required to start.
Access Riven Chat in the browser or download the desktop application for a focused, distraction-free reasoning workspace.
Get API keys, review documentation, and start building on the Riven platform. OpenAI-compatible, with full usage and routing controls.
Riven API is fully OpenAI-compatible. Point your existing code at our endpoint, swap your API key, and you're running on Riven infrastructure. No SDK changes, no rewrites.
SSO, audit logs, access controls, and dedicated support. Self-host if you need to. We'll fill out your security questionnaire.
SSO
SAML and OIDC
Audit logs
Full activity trail
Self-host
Your infrastructure
Access controls
Role-based permissions
Pick the surface that fits your work. Every product connects to the same platform.
Contact sales