Kat Coder Pro V2
Kat Coder Pro V2 is KwaiPilot's second-generation agentic coding model with a context window of 256K tokens and max output of 256K tokens, building on the KAT-Coder series for software engineering tasks.
import { streamText } from 'ai'
const result = streamText({ model: 'kwaipilot/kat-coder-pro-v2', prompt: 'Why is the sky blue?'})Playground
Try out Kat Coder Pro V2 by KwaiPilot. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
About Kat Coder Pro V2
Kat Coder Pro V2 builds on KAT-Coder Pro V1 in the KAT-Coder series. The context window of 256K tokens and maximum output of 256K tokens support tasks involving large codebases or long-form code generation.
Like V1, the model targets agentic software engineering workflows: multi-file editing, issue resolution, and test case generation. It supports parallel tool calling and multi-turn interaction, the same core capabilities that define the KAT-Coder series.
You can route requests through AI Gateway to access Kat Coder Pro V2 via streamlake. Multi-provider support adds resilience for production deployments. See https://www.streamlake.ai/ for documentation from KwaiPilot.
Providers
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
More models by KwaiPilot
| Model |
|---|
| 256K | 3.2s | 106tps | $0.03/M | $1.20/M | Read:$0.06/M Write:— | — | — |  | 10/24/2025 |
What To Consider When Choosing a Provider
- Configuration: Check streamlake for current provider availability. Routing through multiple providers on AI Gateway enables automatic failover. Technical details sit at https://www.streamlake.ai/.
- Zero Data Retention: AI Gateway does not currently support Zero Data Retention for this model. See the documentation for models that support ZDR.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
When to Use Kat Coder Pro V2
Best For
- Large codebase tasks: Software engineering work that benefits from a context of 256K tokens
- Long-form code generation: Outputs need 256K tokens of capacity per response
- Multi-file editing: Refactoring workflows span several files at once
- Multi-provider resilience: Production agent pipelines benefit from routing on AI Gateway
Consider Alternatives When
- V1 covers your needs: KAT-Coder Pro V1 meets context and output requirements at a lower cost
- Multimodal input required: You need image or document input alongside code
- Simple completion model: A lighter, non-agentic completion model suffices for your use case
Conclusion
Kat Coder Pro V2 extends the KAT-Coder series with expanded context and output capacity. For teams that need more headroom than V1 for large codebases or long outputs, it adds that capacity with multi-provider resilience on AI Gateway.
Frequently Asked Questions
What is the context window size of Kat Coder Pro V2?
Kat Coder Pro V2 has a context window of 256K tokens, supporting large codebases and multi-file tasks in a single context.
What is the maximum output token limit?
Up to 256K tokens per request.
When was Kat Coder Pro V2 released?
March 27, 2026.
How does V2 differ from V1?
V2 has a context window of 256K tokens and max output of 256K tokens. Both models share the same core capabilities: parallel tool calling, multi-turn interaction, and coverage of eight software engineering task types.
What is the pricing for Kat Coder Pro V2?
Check the pricing panel on this page for today's numbers. AI Gateway tracks rates across every provider that serves Kat Coder Pro V2.
Which providers serve Kat Coder Pro V2?
The model is available from streamlake through AI Gateway.