UTC

Kimi K2.6

Moonshot AI · open weights · served by 77 providers · tool calling, reasoning, image input, video input

moonshotai/kimi-k2.6

Weights
Open
Released
2026-04-20
Knowledge cutoff
2025-01
Context window
262K
Deprecation dates

When each platform deprecates Kimi K2.6, as LiteLLM records it. Plan the move before the date: after it, the platform may stop serving the model or stop updating it.

  • Azure AI Foundrydeprecates 2027-04-16azure_ai/kimi-k2.6successor: azure_ai/kimi-k3.1-pro
  • Together AIdeprecated 2026-08-19together_ai/moonshotai/Kimi-K2.6successor: together_ai/moonshotai/Kimi-K3.1-Pro
  • Basetendeprecated 2026-10-16baseten/moonshotai/Kimi-K2.6successor: baseten/moonshotai/Kimi-K3.1-Pro
Every upcoming deprecation →
Hugging Face repository

moonshotai/Kimi-K2.6 at commit 7eb5002f6a, last changed 2026-05-19

Licence
modified-mit
Access
Open download
Library
transformers
Formats
.safetensors
  • Loading it runs code from the repository (trust_remote_code). Read that code, and pin the commit.
  • No model signature file is published with the weights.

What the Hub's scanners report

23 weight files, 23 with a scan result. No scanner reports an issue with them.

The model at this commit, its 23 weight files with their SHA-256.

Facts from the Hugging Face Hub and its file scanners (Protect AI, Cisco, Hugging Face picklescan, VirusTotal, JFrog), as of 2026-11-18. ShieldMap does not copy model cards.

Who serves it
  • 302.AIkimi-k2.6
  • Abacusmoonshotai/Kimi-K2.6
  • AIHubMixkimi-k2.6
  • Alibaba Token Plankimi-k2.6
  • Alibaba Token Plan (China)kimi-k2.6
  • Aurikokimi-k2.6
  • Azurekimi-k2.6
  • Azure Cognitive Serviceskimi-k2.6
  • Chutesmoonshotai/Kimi-K2.6-TEE
  • Clarifaimoonshotai/chat-completion/models/Kimi-K2_6
  • ClinePasscline-pass/kimi-k2.6deprecated
  • Cloudflare Workers AI@cf/moonshotai/kimi-k2.6deprecated
  • CommandCodemoonshotai/Kimi-K2.6
  • Cortecskimi-k2.6
  • CrossModelmoonshot/kimi-k2.6
  • Crusoemoonshotai/Kimi-K2.6deprecated
  • Deep Inframoonshotai/Kimi-K2.6
  • DigitalOceankimi-k2.6
  • DotHooshkimi-k2.6
  • EBCloudKimi-K2.6
  • Eden AImoonshot/kimi-k2.6
  • EmpirioLabs AIkimi-k2-6
  • evrocmoonshotai/Kimi-K2.6
  • FastRoutermoonshotai/kimi-k2.6
53 more providers

FrogBot (kimi-k2-6), GMI Cloud (moonshotai/Kimi-K2.6), GreenPT (kimi-k2.6), GreenPT (kimi-k2.6-fast), Hugging Face (moonshotai/Kimi-K2.6), Inceptron (moonshotai/Kimi-K2.6), Infomaniak (moonshotai/Kimi-K2.6), IteraCompute (moonshotai/kimi-k2.6), JZSTOKEN (kimi-k2.6), Kenari (kimi-k2-6), Kilo Gateway (moonshotai/kimi-k2.6), DevPass (LLM Gateway) (kimi-k2.6), LLM Gateway (canopywave/kimi-k2.6), LLM Gateway (moonshot/kimi-k2.6), LLM Gateway (novita/kimi-k2.6), LLM Gateway (tencent/kimi-k2.6), Melious (kimi-k2.6), Merge Gateway (moonshot/kimi-k2.6), Moark (Kimi-K2.6), Moonshot AI (kimi-k2.6), Moonshot AI (China) (kimi-k2.6), NanoGPT (TEE/kimi-k2.6), NanoGPT (moonshotai/kimi-k2.6), NanoGPT (moonshotai/kimi-k2.6:thinking), NEAR AI Cloud (moonshotai/kimi-k2.6), Novita AI (moonshotai/kimi-k2.6), Nvidia (moonshotai/kimi-k2.6), Ofox (moonshotai/kimi-k2.6), Ollama Cloud (kimi-k2.6), OpenCode (kimi-k2.6), OpenRouter (moonshotai/kimi-k2.6), OrcaRouter (kimi/kimi-k2.6), Phala (moonshotai/kimi-k2.6), Poe (novita/kimi-k2.6), Privatemode AI (kimi-k2.6), Privatemode AI (kimi-latest), Requesty (kimi-k2.6), routing.run (kimi-k2.6), routing.run (kimi-k2.6-nitro), Runcrate (moonshotai/Kimi-K2.6), SCNet Token Plan (Kimi-K2.6), SiliconFlow (moonshotai/Kimi-K2.6), SiliconFlow (China) (Pro/moonshotai/Kimi-K2.6), SteadyRouter (kimi-k2.6), TensorX (moonshotai/kimi-k2.6), Thinking Machines (moonshotai/Kimi-K2.6), Thinking Machines (moonshotai/Kimi-K2.6:peft:131072), TokenGo (moonshotai/kimi-k2.6), UnoRouter (kimi-k2.6), Venice AI (kimi-k2-6), Vercel AI Gateway (moonshotai/kimi-k2.6), CoreWeave (moonshotai/Kimi-K2.6), ZenMux (moonshotai/kimi-k2.6)

Running it yourself

Open weights are loaded and served by code that has its own advisories. These inference servers and model loaders are on ShieldMap; load weights only from the publisher's own repository, and prefer safetensors over formats that can run code.

How supply-chain attacks on models work →

Sources: models.dev (MIT) and the LiteLLM model catalogue (MIT). Deprecation dates are as each platform published them and as LiteLLM records them; check the platform before you rely on one. Data as of 2026-11-22.