Kwaipilot: KAT-Coder-Air V2.5

Text input Text output
Author's Description

Key Specifications
Cost
$$$
Context
256K
Released
Jul 10, 2026
Speed
Ability
Reliability
Supported Parameters

This model supports the following parameters:

Stop Max Tokens Logprobs Tools Tool Choice Structured Outputs Response Format Top P Frequency Penalty Presence Penalty Temperature Top Logprobs
Features

This model supports the following features:

Response Format Tools Structured Outputs
Performance Summary

Kwaipilot's KAT-Coder-Air V2.5, created on July 10, 2026, demonstrates exceptional performance across several key metrics. It consistently ranks among the fastest models available and offers highly competitive pricing, placing it in the Infinityth percentile for both speed and cost across eight benchmarks. The model also exhibits outstanding reliability, achieving a 100% success rate across all benchmarks, indicating minimal technical failures. In terms of specific performance, KAT-Coder-Air V2.5 achieves perfect accuracy in General Knowledge and Ethics, making it the most accurate model at its price point and speed in these categories. It also shows strong performance in Mathematics (95.4% accuracy, 89th percentile) and Reasoning (92.0% accuracy, 70th percentile). A notable weakness is its complete failure in Instruction Following (0.0% accuracy), suggesting a significant area for improvement. While its Hallucinations accuracy is moderate (91.7%), its Email Classification (97.0%) and Coding (84.0%) performances are average compared to other models.

Model Pricing

Current Pricing

Feature Price (per 1M tokens)
Prompt $0.15
Completion $0.6
Input Cache Read $0.03

Price History

Available Endpoints
Provider Endpoint Name Context Length Pricing (Input) Pricing (Output)
StreamLake
StreamLake | kwaipilot/kat-coder-air-v2.5-20260710 256K $0.15 / 1M tokens $0.6 / 1M tokens
Benchmark Results
Benchmark Category Reasoning Strategy Free Executions Accuracy Cost Duration
Other Models by kwaipilot