What is Kimi used for?

Kimi 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 Kimi compare to Kimi K2?

Kimi by Moonshot AI is strongest where you need code, while Kimi K2 by Moonshot AI is the closest related family to check for code. Kimi has 1 listed variant and reaches up to 256k context, while Kimi K2 reaches up to 262k context, so compare the specs and pricing tables before choosing a production model.

Which Kimi model should I use?

Kimi K2.5 is both the lowest listed input-price option at $0.44/1M input tokens through OpenRouter and the strongest local starting point with 256k context and function calling and structured outputs. Use the provider table if latency, deployment type, or output-token pricing matters more than input price.

Kimi Models by Moonshot AI

Moonshot AI

1 model2026Up to 256k ctxFrom $0.44/1M input

About

Kimi, created by Moonshot AI, is a pioneering family of large language models renowned for their ability to process extraordinarily long contexts. Initially supporting up to 200,000 Chinese characters (roughly 400 million tokens) 410, its capacity has soared to an impressive 2 million Chinese characters 346, outstripping many leading models like GPT-4's 128,000 tokens 4. This extensive context capability enables Kimi to efficiently summarize long documents, analyze intricate research papers, and carry out complex dialogues. Although primarily designed for the Chinese language, Kimi offers support for several other languages 1410, making it adaptable for applications ranging from scholarly research to software engineering 410. Its continuous evolution is a testament to the swift progress in AI technology 8.

Current Variants

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

1 in view

Kimi K2.5Current

Use when the workload needs code, 256k context, and function calling.

2026-03code256k contextfunction calling

Current Kimi variants with use-when guidance and lifecycle status
Model	Use when	Released	Signals	Status
Kimi K2.5	Use when the workload needs code, 256k context, and function calling.	2026-03	code256k contextfunction calling	Current

Release Timeline

1 release group

2026-03

1 current

Kimi K2.5

code256k contextfunction calling

Current

Specifications(1 models)

Kimi model specifications comparison
Model	Released	Context	Parameters	Fn Calling	Structured Outputs
Kimi K2.5	2026-03	256k	1T (MoE, 384 experts)	Yes	Yes

Available From(10 providers)

AWS Bedrock

Cloudflare Workers AI

Replicate API +2 more

Pricing

Kimi model pricing by provider
Model	Provider	Input / 1M	Output / 1M	Type
Kimi K2.5	OpenRouter	$0.44	$2	Serverless
Kimi K2.5	Together AI	$0.5	$2.8	Serverless
Kimi K2.5	Fireworks AI	$0.6	$3	Serverless
Kimi K2.5	AWS Bedrock	$0.6	$3	Serverless
Kimi K2.5	Replicate API	$0.6	$3	Serverless
Kimi K2.5	Vercel AI Gateway	$0.6	$3	Serverless
Kimi K2.5	Novita AI	$0.6	$3	Serverless
Kimi K2.5	Fireworks AI	$0.99	$4.94	Serverless

Comparisons

All comparisons →

Frequently Asked Questions

What is Kimi used for?: Kimi 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 Kimi compare to Kimi K2?: Kimi by Moonshot AI is strongest where you need code, while Kimi K2 by Moonshot AI is the closest related family to check for code. Kimi has 1 listed variant and reaches up to 256k context, while Kimi K2 reaches up to 262k context, so compare the specs and pricing tables before choosing a production model.
Which Kimi model should I use?: For the lowest listed input price, start with Kimi K2.5 through OpenRouter at $0.44/1M input tokens. For the most capable/latest local choice, evaluate Kimi K2.5 with 256k context and function calling and structured outputs.

Models(1)