Read our launch media release: Dipole ML Inc. officially launches Raven and DCode.
Frontier intelligence that leads the world in terminal execution and complex reasoning, delivered with your data encrypted at rest and never for sale.
Built for engineering teams and autonomous workflows worldwide
Raven Flash and Raven Max share one unified foundation: text and image understanding, native tool calling, and inspectable reasoning that sets the standard on global benchmarks.
Evaluated with maximum reasoning effort enabled. Verified against official frontier model evaluations.
Raven Max · Frontier Reasoning
Our flagship frontier model for complex engineering, long-horizon planning, and rigorous analysis. Raven Max plans across multi-step dependencies, calls tools with precision, and streams extended reasoning so every architectural decision is fully auditable.
Meet Raven MaxRaven Flash · High-Velocity Frontier
Engineered for low latency and high throughput without compromising accuracy. Raven Flash is the default autonomous engine in DCode and the Raven app, delivering real-time responses, tool execution, and code generation at a fraction of standard API latency.
Meet Raven FlashMultimodal Understanding
Attach system architectures, UI mockups, terminal logs, and technical documents. Raven models process images natively alongside code, execute tools within agentic loops, and stream their internal thinking steps in plain view. Nothing happens in an opaque black box.
Explore the family// a Raven conversation, abridged
You: Analyze this Kubernetes cluster architecture diagram and explain why the billing-service pod is crash-looping under peak traffic.
Raven: The failure occurs because the billing worker runs on a shared pool without isolated resource quotas. When the batch job spikes, the node exhausts memory and the OOM killer terminates the billing pod. Move the worker to a dedicated node pool or set explicit limits in deployment.yaml.
We do not sell your content or personal information, share it with advertisers, or run trackers and advertising on our sites or apps. Your prompts power Raven, nothing else.
Saved conversation histories are encrypted at rest using individual Data Encryption Keys (DEK). Each conversation is wrapped under its own key rather than one shared lock, and your history stays encrypted as it syncs across the web, desktop, and mobile apps.
Multi-step tool execution, terminal commands, and file edits in DCode run inside isolated sandboxes. Raven models require explicit human approval before any disk write or command can run, ensuring autonomous workflows remain completely under your control.
Frontier intelligence across every platform. Access Raven Flash and Raven Max instantly through our browser web chat, or install the native app on desktop and mobile. Both provide fluid multimodal understanding, inspectable reasoning streams, and encrypted history.
The autonomous coding environment built for production software engineering. DCode navigates entire codebases, plans multi-step refactors, edits files, executes shell commands, and verifies test suites with receipts. You stay in command of every commit.
Explore DCodeNatively compatible with both OpenAI and Anthropic API formats. Simply point your existing SDK base URL to api.dipoleml.com for sub-second streaming, automatic prompt caching discounts, and pay-as-you-go credits with zero subscription lock-in.
Read the API docsExperience Raven Flash and Raven Max live in your browser. Evaluate multimodal vision, inspectable thinking streams, and tool calling with zero setup or credit card required.
Launch PlaygroundDevelopers
Natively compatible with your existing OpenAI and Anthropic SDKs. Switch your base URL and start streaming in seconds.
from openai import OpenAI # Drop-in replacement: simply update the base_url client = OpenAI( base_url="https://api.dipoleml.com/v1", api_key="your-dipole-api-key" ) response = client.chat.completions.create( model="raven-max", messages=[{"role": "user", "content": "Refactor the auth microservice."}] ) print(response.choices[0].message.content)
# Install the global agentic CLI npm install -g @dipoleml/dcode-cli # Run multi-step agentic refactors directly in your terminal dcode "Audit database connections and run regression suite"
Trust & Governance
Built from the ground up to give organizations complete data autonomy and verifiable security, free from Big Tech cloud lock-in.
Self-operated model serving and routing infrastructure. We run our own gateway and systems, ensuring direct end-to-end provenance without third-party middleman brokers.
Fully compliant with PIPEDA, GDPR, and enterprise compliance standards. Hosted within Canadian jurisdiction, keeping your data protected under strict statutory privacy laws. Read about sovereign AI.
Real free tiers, pay-as-you-go credits, automatic input caching discounts, and zero forced subscriptions. You only pay for the tokens you actually compute.
Works with the coding apps, harnesses, and ADEs engineers already use
News
Launches, benchmarks, and engineering notes from the lab, the moment they land.
Autocomplete, the AI-first editor, or the full agent: a straight comparison of the three philosophies of AI-assisted coding, written by the people who built the third one.
Read the postThe same Raven agent is one npm install away, and the API behind it publishes its price to the cent: $0.50 per 1M input tokens on Flash, credits that never expire, and per-key spend limits.
Read the postChatGPT plugins died in 2024, and the autopsy matters: sandboxed islands with no real account access were never going to work. Connectors, done properly, are the version that survives.
Read the postType "read my last email" into Raven and it does, because connectors give the model real access to your actual accounts through real OAuth consent and tightly scoped read actions.
Read the postThe same Raven, the same chats, everywhere you are: a full web app, native Windows and macOS apps, and Raven on Android and iPhone, syncing live across every screen.
Read the postAt Dipole ML Inc., your data is encrypted at rest, access is limited, and nothing is ever sold. Here is our privacy model in plain terms, plus a two-minute checklist for any AI app.
Read the postRaven's account-level memory carries what matters across every conversation and every device: auto-captured where useful, always visible, and yours to edit or delete at any time.
Read the postAsk Raven for a presentation or a report and the download is a real file: PowerPoint, Word, Excel, PDF, Markdown, or CSV, assembled on your device and ready for the tools you already use.
Read the postAsk Raven a question that needs the live web and it asks permission, reads the pages, and comes back with an answer whose numbered sources you can click and verify yourself.
Read the postThe full reasoning behind our pricing: a real free tier, Pro at $20, Ultra at $100, API credits from $5 that never expire, and annual rates locked lower for Founding Members.
Read the postInvite the whole team into one Raven conversation: everyone types, everyone sees every message the moment it lands, and Raven answers in the middle of the group, live on every screen.
Read the postEvery Raven chat can be shared like a document: private links that stay blurred until you approve them, public links for signed-in readers, and a presence eye that shows who is in the room.
Read the postThe limited launch cohort for Founding Members is open: locked-in first-year annual pricing across Pro, Ultra, and DCode, plus a permanent golden badge.
Read the postSovereign AI is intelligence you can verify end to end: models and infrastructure under your own jurisdiction, with privacy enforced by architecture rather than policy promises.
Read the postBoth Raven models carry a one-million-token context window and 128,000 tokens of output. Here is what that actually holds, and the workflows it changes once the model stops forgetting.
Read the postA technical deep dive into the system surrounding the Raven model family: transparent reasoning streams, encryption at rest, careful access controls, and prompt injection defense.
Read the postThe DipoleML API at api.dipoleml.com speaks both OpenAI and Anthropic SDK formats natively, so migrating is a one-line base URL change with zero code rewrites.
Read the postDCode 1.0 ships for macOS, Windows, and the npm CLI: native MCP support, deep codebase graphs, self-healing shell execution, and Windows computer use.
Read the postRaven Flash takes #1 on AutomationBench (49.2), leads the velocity tier on Terminal Bench 4.0 (34.1), and ties #1 on the Intelligence Index, at $0.50 per 1M input tokens.
Read the postThe full technical evaluation of Raven Max: #1 on Terminal Bench 4.0 (42.8), Agent's Last Exam (39.5), and GDPVal-AA (1,771), with maximum reasoning effort enabled.
Read the postThe official launch media release: Raven Flash, Raven Max, DCode 1.0, and the API gateway are live, built and served from Canada on three pillars: access, sovereignty, and trust.
Read the postFounder Saurabh Sharma announces the October 1, 2026 public launch of Dipole ML Inc.: the Raven model family and DCode, after four years of research.
Read the postExperience autonomous coding, deep reasoning, and verifiable intelligence built for everyone.
No credit card required. Usage replenishes automatically. Switch your stack in minutes.