Kimi K2.5
Moonshot AI · open weights · served by 46 providers · tool calling, reasoning, image input
moonshotai/kimi-k2.5
- Weights
- Open
- Released
- 2026-01-27
- Knowledge cutoff
- 2025-01
- Context window
- 262K
When each platform deprecates Kimi K2.5, 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 2027-01-26azure_ai/kimi-k2.5successor: azure_ai/kimi-k2.6
- moonshotdeprecated 2026-08-31moonshot/kimi-k2.5successor: moonshot/kimi-k3.1-pro
- Basetendeprecated 2026-10-16baseten/moonshotai/Kimi-K2.5successor: baseten/moonshotai/Kimi-K3.1-Pro
moonshotai/Kimi-K2.5 at commit 4d01dfe033, last changed 2026-04-30
- 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
33 weight files, 33 with a scan result. No scanner reports an issue with them.
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.
- 302.AIkimi-k2.5
- Abacuskimi-k2.5deprecated
- AIHubMixkimi-k2.5
- Alibaba Token Plankimi-k2.5
- Alibaba Token Plan (China)kimi-k2.5
- Amazon Bedrockmoonshotai.kimi-k2.5
- Aurikokimi-k2.5
- Azurekimi-k2.5
- Azure Cognitive Serviceskimi-k2.5
- CommandCodemoonshotai/Kimi-K2.5
- DaoXEkimi-k2.5
- Deep Inframoonshotai/Kimi-K2.5deprecated
- Eden AIamazon/moonshotai.kimi-k2.5
- Eden AIdeepinfra/moonshotai/Kimi-K2.5
- Eden AItensorx/moonshotai/kimi-k2.5
- FrogBotkimi-k2.5deprecated
- HPC-AImoonshotai/kimi-k2.5
- Hugging Facemoonshotai/Kimi-K2.5
- IteraComputemoonshotai/kimi-k2.5
- Jalapeno CloudKimi-K2.5
- Jiekou.AImoonshotai/kimi-k2.5
- Kilo Gatewaymoonshotai/kimi-k2.5
- DevPass (LLM Gateway)kimi-k2.5
- LLM Gatewayalibaba/kimi-k2.5
22 more providers
LLM Gateway (moonshot/kimi-k2.5), Meganova (moonshotai/Kimi-K2.5), Merge Gateway (moonshot/kimi-k2.5), NanoGPT (moonshotai/kimi-k2.5), NanoGPT (moonshotai/kimi-k2.5:thinking), Novita AI (moonshotai/kimi-k2.5), OpenCode (kimi-k2.5), OpenCode (kimi-k2.5-free), OpenRouter (moonshotai/kimi-k2.5), OrcaRouter (kimi/kimi-k2.5), Poe (fireworks-ai/kimi-k2.5-fw), Poe (novita/kimi-k2.5), Qiniu (moonshotai/kimi-k2.5), Requesty (kimi-k2.5@eu), SCNet Token Plan (Kimi-K2.5), SiliconFlow (moonshotai/Kimi-K2.5), SiliconFlow (China) (Pro/moonshotai/Kimi-K2.5), Tencent Coding Plan (China) (kimi-k2.5), TensorX (moonshotai/kimi-k2.5), Vercel AI Gateway (moonshotai/kimi-k2.5), Watt Cloud (Kimi-K2.5), Zenifra (zenifra/kimi-k2.5)
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.
- vllm · 131 advisories
- transformers · 24 advisories
- keras · 20 advisories
- onnx · 17 advisories
- bentoml · 16 advisories
- github.com/ollama/ollama · 16 advisories
- torch · 13 advisories
- lmdeploy · 10 advisories
- sglang · 10 advisories
- skops · 4 advisories
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.