Context
256K
Max output
128K
Serving providers
1
Cheapest input
$0.30 /1M
Prices are on-demand list rates. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.

Benchmarks

independent evaluations · source: Artificial Analysis
Speed & latency
Output speed
98 t/s
tokens / second
Time to first token
6.47s
latency
Headline indices
59.5
of 100
15.5
of 100
Individual benchmarks · click a row for effort variants
BenchmarkBest scoreEffortCost / task
CritPt0%default
GDPval908.2default
GPQA Diamond85%default
Humanity's Last Exam16%default
IFBench67%default
Long-Context Reasoning66%default
Omniscience-21.6default
SciCode38%default
TerminalBench Hard49%default
TerminalBench v2.170%default
τ-bench Banking6%default
τ²-bench89%default

Pricing by serving provider

standard tier · per 1M tokens · click a provider to expand
Serving providerInput /1MOutput /1MEndpoints
NovitaCheapest$0.30$1.201
ComponentUnitStandardBatchCached
Text cached in/1M tok$0.0600
Text input/1M tok$0.3000
Text output/1M tok$1.20

Price history

input + output $/1M since we started tracking
Input Output
$0.000$0.500$1.00$1.50Jan 13Jun 22$1.20$0.300

Cost calculator

estimate your monthly spend on this model
$156
estimated / month

Model IDs

copy the exact identifier for your platform
novita/kwaipilot/kat-coder-pro

Frequently asked questions

KAT-Coder-Pro V2 pricing, context and availability

How much does KAT-Coder-Pro V2 cost?

KAT-Coder-Pro V2 costs $0.30 per 1M input tokens and $1.20 per 1M output tokens at its cheapest provider via Novita.

What is the context window of KAT-Coder-Pro V2?

KAT-Coder-Pro V2 accepts up to 256K tokens of context and can return up to 128K output tokens.

What can KAT-Coder-Pro V2 do?

KAT-Coder-Pro V2 supports tool use and structured output.

Other Other models

compare pricing across the Other lineup
Pricing verified against LiteLLM + OpenRouter + provider pages · refreshed daily

Every weekday

AI moves fast. Here's your debrief.

News, analysis, tools, and more.

For people who build with AI