Kwaipilot Releases KAT-Coder-Pro V2.5 with 256K Context Window at $0.74/$2.96 Per Million Tokens
Kwaipilot has released KAT-Coder-Pro V2.5, a coding-focused language model with a 256,000-token context window. The model is priced at $0.74 per million input tokens and $2.96 per million output tokens, available through OpenRouter.
KAT-Coder-Pro V2.5 — Quick Specs
Kwaipilot Releases KAT-Coder-Pro V2.5 with 256K Context Window
Kwaipilot has released KAT-Coder-Pro V2.5, a coding-focused language model with a 256,000-token context window. The model is priced at $0.74 per million input tokens and $2.96 per million output tokens.
The model is available exclusively through OpenRouter, which forwards requests directly to Kwaipilot without routing decisions. According to OpenRouter's data, prompt caching can reduce effective costs by 60-80% below list prices depending on the amount of repeated context.
Technical Specifications
- Context Window: 256,000 tokens
- Input Pricing: $0.74 per 1 million tokens
- Output Pricing: $2.96 per 1 million tokens
- Release Date: July 10, 2026
- Model Type: Text (coding-focused)
- API: OpenAI-compatible through OpenRouter
The model slug is kwaipilot/kat-coder-pro-v2.5 for API integration. OpenRouter provides monitoring data including throughput (tokens per second), latency, time-to-first-token (TTFT), and 30-day uptime metrics for the model.
Pricing Context
The output pricing of $2.96 per million tokens positions KAT-Coder-Pro V2.5 in the mid-range for coding models. The input cost of $0.74 per million tokens is competitive for models with 256K context windows.
OpenRouter's infrastructure automatically retries failed requests on alternative providers and tracks which applications send the most traffic to the model, providing signals about production workload patterns.
What This Means
The 256K context window allows developers to include substantial codebases or documentation in prompts without chunking. The pricing structure makes it viable for production coding assistance applications, though the 4:1 output-to-input price ratio means code generation workloads will cost significantly more than code analysis. The exclusive availability through OpenRouter rather than direct API access may limit adoption compared to models with multiple distribution channels.
Related Articles
Microsoft releases Decision-1, a Qwen3.5-9B-based model for classification and routing, at $0.042 per 1M input tokens
Microsoft has released Decision-1, a decision model built on Qwen3.5-9B for classification, evaluation, and routing. Microsoft claims 83.5% accuracy across 36 benchmarks and 85 ms latency. Input tokens cost $0.042 per million, and output tokens are free.
StepFun releases Step 5 Preview: 600B MoE with 1M context at $1/$2.70 per 1M tokens
StepFun has listed Step 5 Preview, a sparse Mixture-of-Experts model with 600B total and 27B active parameters and a 1.0M-token context window. It is priced at $1 input and $2.70 output per 1M tokens on OpenRouter. StepFun positions it as its flagship model for agentic work.
Google releases Nano Banana 2.1 image model: $1.50/$30 per 1M tokens, 66K context
Google's Nano Banana 2.1 (Gemini Nano Banana 2.1) is an image generation and editing model on the Flash tier, listed on OpenRouter at $1.50 input and $30 output per 1M tokens with a 66K context window. It supports 1K, 2K, and 4K output and succeeds Nano Banana 2 and Nano Banana Pro, according to the listing.
Odyssey-3 world model enters free public preview; 14B base model, 66.1 Physics-IQ claim carries caveats
Odyssey has opened a free public research preview of Odyssey-3, an autoregressive diffusion transformer that generates interactive environments from text prompts in real time. The base model has 14 billion parameters and outputs 832 × 480 video. Odyssey's headline Physics-IQ score of 66.1 does not meet the benchmark's own requirements for a record claim.
Comments
Loading...