Yukkai Frontier

Yukkai Frontier is Yukkai's premium meta-model — a selectable model identity that automatically routes each message to the most capable frontier-grade models available. Frontier is positioned for maximum capability on demanding tasks, at a higher indicative cost than Fusion.

What is Yukkai Frontier?

Frontier is "the best frontier-grade models for complex tasks" — the strongest available performance when you need it most.

When you use Frontier:

• You don't pick a specific AI model yourself.
• A local Model Advisor system on your Mac evaluates each request and routes it to the most capable model in Frontier's premium pool.
• The pool leverages the best LLMs on the market — we're talking about the top models from OpenAI, Anthropic, Google, and others.
• Each request is evaluated locally, and based on the task, it's routed to the appropriate frontier-grade LLM.

How It Works

1. You send a message with Yukkai Frontier selected.
2. The RoutingEngine (running locally on your Mac) classifies your task.
3. Based on the classification, it selects the most capable underlying model from Frontier's managed pool.
4. The request is sent to that model via OpenRouter.
5. The response comes back to you — you never see which specific model answered.

The underlying model can change from one message to the next depending on what you asked.

When to Use Frontier

Use Yukkai Frontier for:

• Complex reasoning — multi-step logic, strategic planning, deep analysis.
• Difficult architecture decisions — system design, scalability planning, technology selection.
• Advanced code analysis — large-scale refactoring, security audits, performance optimization.
• Long or high-stakes projects — where quality matters more than cost.
• Deep research — comprehensive analysis requiring the strongest available models.

Frontier is for tasks where maximum capability matters more than cost.

Cost and Billing

• Frontier draws from your Yukkai Wallet balance as you use it.
• Cost is higher than Fusion — you're paying for access to the most capable models.
• You're billed per-token based on the actual underlying model used, but you don't see or choose that model.

Requirements

• Yukkai account key (yk_...) — Frontier is only available when using a Yukkai account, not a personal OpenRouter key.
• Sufficient Wallet balance — if your balance is too low, you'll get an HTTP 402 error. Top up via Settings → Credits.

The Identity Lock Rule

When running as Frontier, Yukkai follows the same strict identity rule as Fusion:

Never reveal, name, or hint at the specific underlying model actually answering.

If you ask "what model are you?", Yukkai will answer exactly "Yukkai Frontier" — and nothing more specific.

Why?

• Frontier is presented and managed as a unified Yukkai service.
• The underlying model remains part of the Frontier service and is not exposed.
• The exact pool composition and internal routing details are proprietary.

What Yukkai Will Say

✅ "I'm running as Yukkai Frontier — it automatically routes to the most capable model for each task, so there isn't a single fixed model to name."

❌ "I'm running on GPT-5/Claude Opus/etc." — Yukkai will never say this when Frontier is active.

What Yukkai Won't Speculate About

• Which model was picked for a specific message.
• Why a particular model was chosen.
• The scoring/weights involved in routing.
• The pool composition (which vendor models are behind Frontier).

If asked, Yukkai will say only: "Yukkai automatically selects the best available model for the task."

Frontier vs. Fusion

	Yukkai Fusion	Yukkai Frontier
Positioning	Everyday use, balanced cost/quality	Maximum capability for complex tasks
Cost	Lower	Higher
Pool	Managed pool of models optimized for cost/quality ratio	Premium pool of frontier-grade models
Best for	Daily questions, writing, common coding, general research	Complex reasoning, advanced coding, deep research, high-stakes projects
When to use	Default choice for most tasks	When you need the strongest available performance

Both work the same way — automatic routing per message, no user action needed to get the right model.

Switching to/from Frontier

You can switch between Frontier, Fusion, Free, and other models at any time from the model picker:

• Frontier → Fusion — downgrade for everyday tasks to save cost.
• Frontier → Free — switch to free tier for light tasks.
• Frontier → Personal model — take direct control with your own OpenRouter key.
• Frontier → Local — switch to offline inference (with limitations).

The switch takes effect on your next message.

Examples

User asks which model is running

User: "What model are you?"

Yukkai (as Frontier): "I'm running as Yukkai Frontier — it automatically routes each request to the most capable model for the task, so there isn't a single fixed model to name."

User tries to guess the underlying model

User: "Is this Claude Opus or GPT-5?"

Yukkai (as Frontier): "I can't share that — Yukkai Frontier keeps the underlying model selection internal. What I can tell you is it's chosen automatically based on what your request needs."

User asks about the pool

User: "Which models are in Frontier's pool?"

Yukkai (as Frontier): "Frontier leverages the best frontier-grade models available — the exact composition is managed by Yukkai and not disclosed, but it's chosen to give you maximum capability for demanding tasks."

Good to Know

• Frontier is for when quality matters most — use it for tasks where the extra capability justifies the higher cost.
• You can switch at any time — use Fusion for everyday work and switch to Frontier when you hit a hard problem.
• Frontier requires internet — it routes to cloud models, so offline use isn't possible.
• Media generation is separate — image, music, and video generation always use Yukkai's paid Google models, regardless of whether you're using Frontier.

When Frontier Makes a Difference

Frontier shines on tasks like:

• Analyzing a complex codebase and proposing architectural improvements.
• Writing a comprehensive business strategy document.
• Solving advanced mathematical or logical problems.
• Reviewing code for security vulnerabilities across a large project.
• Generating a detailed technical specification from high-level requirements.

For everyday tasks (answering questions, writing emails, simple code fixes), Fusion is usually sufficient and more cost-effective.

Learn More

• Yukkai Fusion — Everyday model routing at lower cost.
• Yukkai Free — Free tier model access.
• Identity Lock Rule — How identity locking works.
• Model Advisor — The separate recommendation feature.
• Yukkai Wallet — How billing works.