LLM ReferenceLLM Reference

KAT Models

4 models2025–2026Up to 256K ctxFrom $0.3/1M input

About

KAT is a family of 4 AI models, released between 2025 and 2026.

Current Variants

Use-when guidance is derived from seed capabilities, context, release, and replacement fields.

4 in view

Use when the workload needs code, 256K context, and tool use.

2026-03code256K contexttool use
KAT CoderCurrent

Use when the workload needs code and 256K context.

2025-01code256K context

Use when the workload needs 128K context and 32B parameters.

2025-01128K context32B parameters

Use when the workload needs 128K context and 72B parameters.

2025-01128K context72B parameters

Release Timeline

2 release groups
2026-03
1 current
KAT Coder Pro V2
code256K contexttool use
Current
2025-01
3 current
KAT Coder
code256K context
Current
KAT Dev 32B
128K context32B parameters
Current
KAT Dev 72B Exp
128K context72B parameters
Current

Specifications(4 models)

KAT model specifications comparison
ModelReleasedContextParametersFn CallingTool UseStructured OutputsCode Exec
KAT Coder Pro V22026-03256KYesYesYesYes
KAT Coder2025-01256KNoNoNoNo
KAT Dev 32B2025-01128K32BNoNoNoNo
KAT Dev 72B Exp2025-01128K72BNoNoNoNo

Available From(2 providers)

Pricing

KAT model pricing by provider
ModelProviderInput / 1MOutput / 1MType
KAT Coder Pro V2OpenRouter$0.3$1.2Serverless
KAT CoderFireworks AI$0.9$0.9Serverless
KAT Dev 32BFireworks AI$0.9$0.9Serverless
KAT Dev 72B ExpFireworks AI$0.9$0.9Serverless

Frequently Asked Questions

What is KAT used for?
KAT is used for code, agent workflows and tool use, and structured outputs. The family description and listed model capabilities point to those workloads as the best fit.
How does KAT compare to Claude 3?
KAT is strongest where you need code, while Claude 3 by Anthropic is the closest related family to check for vision and multimodal work. KAT has 4 listed variants and reaches up to 256K context, while Claude 3 reaches up to 200K context, so compare the specs and pricing tables before choosing a production model.
Which KAT model should I use?
For the lowest listed input price, start with KAT Coder Pro V2 through OpenRouter at $0.3/1M input tokens. For the most capable/latest local choice, evaluate KAT Coder Pro V2 with 256K context and tool use, function calling, and structured outputs.

Models(4)