Moonshot AI

Kimi K2.5

Generally availableOpen weights262K contextReasoning

Summary

Kimi K2.5 is an open-weight language model from Moonshot AI, released on 27 Jan 2026. It accepts text and images and generates text, with a 262K-token context window and up to 16K output tokens per response. As of 18 Nov 2026, the lowest listed API price is $0.45 per million input tokens and $2.25 per million output tokens (SiliconFlow), across 7 providers we track. It is scheduled for retirement on 18 Nov 2026.

Earlier Kimi frontier model for long-context agents, coding, and multimodal work

Last verified 18 Nov 2026. Every figure on this page links to its source.

Specifications

DeveloperMoonshot AI
Release date27 Jan 2026 [models.dev]
StatusGenerally available [models.dev]
WeightsOpen weights [Hugging Face]
Licensemodified-mit [Hugging Face]
Parameters1T [Hugging Face]
Context window262K (262,144 tokens) [models.dev]
Max output16K tokens [models.dev]
Knowledge cutoffJan 2025 [models.dev]
InputText and images [models.dev]
OutputText [models.dev]
Reasoning modeYes [models.dev]
Tool callingYes [models.dev]
Structured outputYes [models.dev]
Retirement date18 Nov 2026 [OpenRouter]
Hugging Facemoonshotai/Kimi-K2.5 [Hugging Face]

Kimi K2.5 API pricing by provider

USD per million tokens. Sorted by input price.

ProviderInputOutputCached inputContextSource
SiliconFlow
moonshotai/Kimi-K2.5
$0.45$2.25$0.07262Kmodels.dev, 18 Nov 2026
OpenRouter
moonshotai/kimi-k2.5
$0.50$2.50$0.15262KOpenRouter, 18 Nov 2026
Azure AI Foundry
kimi-k2.5
$0.60$3–262Kmodels.dev, 18 Nov 2026
Amazon Bedrock
moonshotai.kimi-k2.5
$0.60$3–262Kmodels.dev, 18 Nov 2026
Novita AI
moonshotai/kimi-k2.5
$0.60$3$0.10262Kmodels.dev, 18 Nov 2026
Hugging Face Inference Providers
moonshotai/Kimi-K2.5
$0.60$3$0.10262Kmodels.dev, 18 Nov 2026
Vercel AI Gateway
moonshotai/kimi-k2.5
$0.60$3–256Kmodels.dev, 18 Nov 2026

Running Kimi K2.5 locally

Estimated memory, from the parameter count and the architecture in config.json on Hugging Face. Includes about 0.8 GB of runtime overhead.

QuantizationWeightsTotal, 8K contextTotal, 32K context
Q4_K_M (4-bit)623 GB624 GB626 GB
Q8_0 (8-bit)1091 GB1092 GB1094 GB
FP16 / BF16 (16-bit)2054 GB2055 GB2057 GB

At 4-bit with an 8K conversation, Kimi K2.5 needs about 624 GB, more than any single consumer graphics card; it needs a unified-memory machine, a multi-GPU setup or a split between graphics card and system RAM. Check your own hardware, or see Best computer for local AI on our sister site AIPCs.

Frequently asked questions

What is the context window of Kimi K2.5?

Kimi K2.5 has a context window of 262,144 tokens (262K) and can generate up to 16,384 tokens in a single response, according to models.dev.

How much does the Kimi K2.5 API cost?

Kimi K2.5 is listed from $0.45 per million input tokens and $2.25 per million output tokens (SiliconFlow). Prices last verified 18 Nov 2026.

Is Kimi K2.5 open source?

Kimi K2.5 is an open-weight model: its weights are published on Hugging Face as moonshotai/Kimi-K2.5 under the modified-mit license. Check the license terms before commercial use.

When was Kimi K2.5 released?

Moonshot AI released Kimi K2.5 on 27 Jan 2026, according to models.dev.

Which providers offer Kimi K2.5?

We track Kimi K2.5 on 7 providers: SiliconFlow, OpenRouter, Azure AI Foundry, Amazon Bedrock, Novita AI, Hugging Face Inference Providers, Vercel AI Gateway.

How much VRAM does Kimi K2.5 need?

About 624 GB at 4-bit (Q4_K_M) with an 8K-token conversation, and about 1092 GB at 8-bit. That is more than any single consumer graphics card holds. These are estimates from the model's published parameter count and architecture.

Similar models from other labs

Comparable on output type, weights, release date, price and context window. Prices are the lowest listed, USD per million tokens.

ModelLabReleasedContextInput $/1MOutput $/1MWeights
Qwen3.5 122B-A10BAlibaba (Qwen)23 Feb 2026262K$0.26$2.08Open
Qwen3.5 27BAlibaba (Qwen)23 Feb 2026262K$0.25$2Open
GLM-5Zhipu AI (Z.ai)12 Feb 2026205K$0.95$2.55Open
GLM-4.7Zhipu AI (Z.ai)22 Dec 2025205K$0.40$1.75Open
Step 3.7 FlashStepFun29 May 2026256K$0.20$1.15Open
Macaron V1 TallMindlab21 Jul 2026262K$0.25$1.43Open

Change history

  • Max output changed from 262K to 16K
  • Input modalities changed from text, image, video to text, image
  • Name changed from Kimi-K2.5 to Kimi K2.5
  • Name changed from Kimi K2.5 to Kimi-K2.5
  • Name changed from moonshotai/Kimi-K2.5 to Kimi K2.5
  • Context window changed from 262K to 262K
  • Input modalities changed from text, image to text, image, video
  • Max output changed from 262K to 262K
  • Context window changed from 262K to 262K
  • Input modalities changed from text, image, video to text, image
  • Max output changed from 262K to 262K
  • Name changed from Kimi K2.5 to moonshotai/Kimi-K2.5
  • Max output changed from 16K to 262K
  • Input modalities changed from text, image to text, image, video
  • Context window changed from 256K to 262K
  • Max output changed from 256K to 16K
  • OpenRouter price changed: input $0.53 → $0.50, output $3.33 → $2.50 per 1M tokens
  • Context window changed from 262K to 256K
  • Max output changed from 262K to 256K
  • Max output changed from 256K to 262K
  • Context window changed from 256K to 262K
  • Context window changed from 262K to 256K
  • Max output changed from 262K to 256K
  • Name changed from moonshotai/Kimi-K2.5 to Kimi K2.5
  • OpenRouter price changed: input $0.50 → $0.53, output $2.50 → $3.33 per 1M tokens

Sources