Skip to main content
TokenCost logoTokenCost

KAT-Coder-Pro v2.5

Last verified: July 16, 2026

$0.740/1M input · $2.96/1M output · 262K context · Kwaipilot

Count Tokens

—
Tokens
—
Input Cost
—
Output Cost

Estimate Monthly Cost

Monthly Cost Estimator

Quick:
< $0.0001/mo

Pick a preset above or enter custom usage

Alternatives to KAT-Coder-Pro v2.5

Cheaper Alternative
Kimi K2.5
$0.600 in · $3.00 out · Moonshot
3% cheaper with same or larger context
Premium Upgrade
Grok 4.3
$1.25 in · $2.50 out · xAI
Upgrade with 1000K context
Same Provider
KAT-Coder-Air v2.5
$0.150 in · $0.600 out · Kwaipilot
80% cheaper in Kwaipilot ecosystem

Pricing Details

KwaipilotProprietary, API-only coding model from Kwaipilot (the AI team at Kuaishou), served via the StreamLake platform, OpenRouter, and other aggregators. Cached input $0.15/1M. Kwaipilot's arXiv report (2607.05471) self-reports SWE-Bench Pro 65.2 (2nd behind Opus 4.8's 69.2), PinchBench 94.9 (leads Opus's 93.5), but Terminal-Bench 2.1 only 60.7 (vs Opus 84.6, GLM-5.2 77.9). Report uses SWE-Bench Pro, not the Verified set older KAT versions quoted. No independent eval yet. Distinct from Kwaipilot's open-weight KAT-Dev line on Hugging Face.
Input / 1M tokens
$0.74
Output / 1M tokens
$2.96
Context Window
262K
Max Output
80K
Full Pricing TableToken Calculator

Price History

Launched at current price on 2026-07-10. No price changes recorded.

Frequently Asked Questions

Common questions about KAT-Coder-Pro v2.5 pricing and usage

Read More About KAT-Coder-Pro v2.5

A Kuaishou team just priced 94 percent of Opus 4.8's coding score at a seventh of the cost. The two things holding KAT-Coder v2.5 back are that you can't download it and it stumbles in the terminal.
July 16, 2026 · 8 min read
Full Pricing TableCompare ModelsCost Calculator