Kimi K2 Thinking by Merge Gateway | AI model information

Model explorer

Models

Subscribe

Follow catalog updates and new guides

Open RSS feed Medium guides

RSS covers catalog changes. Medium covers how-to guides and write-ups.

Kimi K2 Thinking by Merge Gateway | AI model information

The fastest GLM 5.2, Kimi K2.6 and MiniMax M3The fastest GLM 5.2, Kimi K2.6 and MiniMax M3 Discuss

Model details

Kimi K2 Thinking

Loading model overview

Merge Gatewaymoonshotai/kimi-k2-thinkingkimi-thinking

Open provider page Provider docs

Quick Info

Provider: Merge Gateway
Model key: moonshotai/kimi-k2-thinking
Release date: Nov 6, 2025

Cost

Input token cost: $0.60
Output token cost: $2.50

Limits

Output tokens: 262,144 tokens
Context window

Latest news about Kimi K2 Thinking

Videos about Kimi K2 Thinking

More models around Kimi K2 Thinking

Also served by

This exact model name is also listed by 20 other providers.

Amazon Bedrock Azure Azure Cognitive Services Helicone Hugging Face IO.NET Kimi For Coding LLM Gateway Meganova Moonshot AI Moonshot AI (China)NanoGPT NovitaAI Ollama Cloud OpenCode Zen OpenRouter Poe Synthetic Vercel AI Gateway Vertex

3 providers

Currently listed through these providers:

Cortecs Fireworks AI Kimi For Coding

3 providers

Kimi K2 Thinking Turbo

Currently listed through these providers:

LLM Gateway Moonshot AI Moonshot AI (China)

2 providers

Currently listed through these providers:

Cortecs Kimi For Coding

2 providers

moonshotai/Kimi-K2-Thinking

Currently listed through these providers:

SiliconFlow SiliconFlow (China)

1 provider

Currently listed through:

1 provider

Kimi K2.6 Turbo

Currently listed through:

1 provider

Kimi K2.6 Thinking

Currently listed through:

1 provider

Kimi K2.5 Thinking TEE

Currently listed through:

1 provider

Kimi K2.5 Thinking

Currently listed through:

1 provider

Pro/moonshotai/Kimi-K2-Thinking

Currently listed through:

SiliconFlow (China)

1 provider

Kimi K2 Thinking Original

Currently listed through:

1 provider

Kimi K2 Thinking Turbo Original

Currently listed through:

Discuss this model

Powered byHyvor Talk

Add corrections, implementation notes, pricing changes, or usage caveats for other readers.

Last updated

Nov 6, 2025

Knowledge cutoff

2024-08

Input modalities

Output modalities

Capabilities

262,144 tokens