Kwaipilot: KAT-Coder-Air V2.5
Kwaipilot's KAT-Coder-Air V2.5 is an AI model designed for handling text inputs within an expansive context length of 256,000 tokens and supports the integration of tools, making it versatile for complex tasks that require external data interaction. While it does not offer reasoning or structured output capabilities, this makes it more suited for scenarios where processing large volumes of textual information is key. Given its pricing at $0.15 per million input tokens and $0.6 per million output tokens, Kwaipilot's model may be a cost-effective choice for projects that primarily focus on text-heavy applications; however, until independent benchmarks are available, potential users should carefully evaluate its performance against specific needs.
- Model ID
- kwaipilot/kat-coder-air-v2.5
- Vendor
- kwaipilot
- Released
- July 2026
- Tokenizer
- Other
- Input Modalities
- text
- Output Modalities
- text
- Max Output
- 80,000 tokens
- Tool Calling
- ✓ supported
- Structured Output
- ✓ supported
- Reasoning Mode
- not supported
- Vision
- text only
- Audio
- no
- Moderated
- no
What it costs in practice
Computed from the current $0.15/M input and $0.60/M output rates. Run your own numbers →
| Job | Tokens | Cost |
|---|---|---|
| Summarize a 50-page report | 30k in / 1.5k out | under $0.01 |
| Classify 1,000 customer emails | 500k in / 50k out | $0.10 |
| A month of a busy support chatbot | 5M in / 2M out | $1.95 |
Similar models
Quick answers
- How much does Kwaipilot: KAT-Coder-Air V2.5 cost?
- $0.15 per million input tokens and $0.60 per million output tokens.
- What is Kwaipilot: KAT-Coder-Air V2.5's context window?
- 256,000 tokens, roughly 384 pages of text in a single request.
- Does Kwaipilot: KAT-Coder-Air V2.5 support tool calling?
- It supports tool calling, structured output.
- Can Kwaipilot: KAT-Coder-Air V2.5 process images?
- No, it is text-only on the input side.