- Models: Direct access to supported frontier, everyday, and lite models from providers like OpenAI, Anthropic, Google, xAI, Moonshot, and Qwen.
- Model Routers: Kombai’s internal routers that automatically pick the ideal model for your task within a chosen cost tier.
The credit multiplier is relative to the baseline (
1x). For example, a task
that consumes 10 credits on a 1x model would consume approximately 5 credits
on a 0.5x model and 25 credits on a 2.5x model.Models
If you prefer to run a specific model, choose one directly. The tables below group models by provider and show each one’s credit multiplier.OpenAI
Anthropic
Gemini
xAI
Moonshot
Qwen
MiniMax
Xiaomi
Meta
Z.ai
Model Routers
Model Routers are powered by a continuously optimized and benchmarked model stack. Kombai benchmarks the latest LLMs from top providers across a wide range of frontend tasks—from interpreting complex UI logic to multi-file refactors—and the internal router automatically selects the best model for the task within the cost tier you choose. This gives you granular control over cost while ensuring you consistently get the best performance per dollar, even as newer and better models emerge, without having to manually test and benchmark them.Kombai Auto is the recommended default. It analyzes each task and routes
it to the optimal model automatically, balancing quality and cost for you.
Because it can pick a different model for each turn, its credit multiplier is
a range:
0.1x and 1x.Select a Model or Model Router
- Click the Model dropdown in the chat input box.
- Search or sort the list to quickly find the Model or Model Router you want to use.
- Click a Model or Model Router to select it.
Sort the Model list
The Sort by dropdown includes the following options:- Recommended: The default Kombai-recommended order.
- Frontier first: Flagship models first, followed by standard and lite models.
- Cost optimized first: Lowest-cost and discounted models first.
- Newest first: Sorted by release date.
- Model routers first: Model Routers first, followed by individual models.
- Alphabetical (A to Z): Alphabetical by model name.
Badges
A badge may appear in each row in the list to indicate whether the item is a router, recommended, new, or cost-optimized.Model details
Each Model and Model Router has a detail view that opens when you hover over it in the list, or when you click Edit (in the IDE extension). The detail view shows the following information:- Type: The model tier (Frontier, Everyday, or Lite).
- Cost: The credit tier (Low, Medium, or High) along with the credit multiplier.
- Context: The model’s context window size. Routers omit this because the active model varies per turn.
- Thinking Effort: The thinking effort levels available for that model.
Thinking effort
Thinking effort controls how deeply a model reasons before it responds. Higher effort improves planning, edge-case handling, and self-correction for complex tasks, while lower effort returns faster, cheaper responses for simpler ones.
Not every model supports every level. When you open the thinking effort selector for a model, only the levels it supports are shown. Some models don’t expose a configurable thinking effort.
Higher thinking effort increases generation time and cost, but it does not
consume your context window, allowing the agent to think deeply without losing
track of large codebases.
Change thinking effort
In the desktop app
In the desktop app
- Click the Model dropdown in the chat input box.
- Hover over a model row (or use the arrow keys) to open its detail card on the side.
- Under Thinking Effort, click the reasoning level you want to use.
In the IDE extension
In the IDE extension
- Click the Model dropdown in the chat input box.
- Hover over a model row and click Edit to open its detail view in place.
- Under Thinking Effort, click the reasoning level you want to set.
- Click All models and routers to return to the list.