Are the weights open and can I run Kimi K2.5 locally?

Yes, the weights are published on Hugging Face at moonshotai/Kimi-K2.5. With 1T total parameters and 32B activated per token via MoE, you will need substantial GPU memory to run it locally — check the VRAM guides before committing.

What does Kimi K2.5 cost and what is its context window?

Via API it is $0.50 per million input tokens and $2.50 per million output tokens, with a 256K token context window.

How is Kimi K2.5 different from Kimi K2?

K2.5 is built on top of Kimi K2 Base with continued pretraining on approximately 15 trillion mixed visual and text tokens, adding native multimodality, visual coding capability, and the agent swarm execution paradigm that K2 lacked.

Kimi K2.5

5202°

#llm #moonshotai #vision #reasoning #agents #coding

Vision

Runs on Graydient

Run this model

Offers / Ads

Easy Creator (Prism) Web

Unlimited · flat fee

Best for beginners. Describe what you want in a chat and the picture comes back, no learning curve.

PirateDiffusion Telegram

Unlimited · flat fee

Images and video from a chat, with short commands. The most convenient tool for Telegram users, with a web companion.

Pro studio (SpyGlass) Web

Unlimited · flat fee

Best for power users. Full control over models, settings and workflows in a browser studio.

Discord bot (Vector)

Unlimited · flat fee

Best for communities. Add it to your own server and run any workflow from a channel.

Unified Editor Web

Unlimited · flat fee

One screen for every model and workflow: prompts, references, edits and video, built into your Graydient account.

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

Local apps

Requires setup

Set up local programs if your computer has capable hardware.

ModelDuel Web

Pay per use

Generate the same prompt on up to 50 checkpoint/base models at a time.

Rent a GPU on the cloud

Graydient Cloud GPU

Unlimited · flat fee

Flat fee, unlimited use. This model installs with one click.

RunPod Cloud GPU

Charged by the minute

Rent a GPU by the minute and set the model up yourself.

Guides for this model

Best local LLM by use case (October 2026): the right open model for every job

How much VRAM do I actually need for local AI?

Open weights vs open source vs API-only: what you are actually allowed to do

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

MoonshotAI | 262,144 token context | $0.50/M input, $2.50/M output. Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed visual and text tokens, it delivers strong performance in general reasoning, visual coding, and agentic tool-calling.

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Kimi K2.5 is Moonshot AI's open-weight multimodal model, built on a 1T-parameter MoE architecture with 32B activated parameters and trained on roughly 15 trillion mixed visual and text tokens. It stands out for native vision-language integration, visual coding from UI designs or video workflows, and a self-directed agent swarm that decomposes complex tasks across parallel specialist agents. It suits developers who need a capable agentic reasoning model they can run themselves.

Known for

Generating code directly from UI screenshots, diagrams, or video workflows
Agent swarm mode: decomposes tasks into parallel domain-specific sub-agents
Strong OCR, document parsing, and chart reasoning on visual inputs
256K token context with both instant and thinking modes available
Open weights on Hugging Face; 1T total parameters, 32B activated per token

Watch out for

1T total parameters means serious hardware requirements to run locally; most users will hit an API.
Licence terms are not detailed in the published card — check the repo before any commercial deployment.
Thinking mode adds latency and output tokens; costs can climb fast on long agentic runs at $2.50/M output.

Best for

Prompts to try

Code from a UI screenshot

I'm attaching a screenshot of a web dashboard. Write clean, production-ready React and Tailwind CSS code that reproduces this layout exactly, including the chart, the sidebar navigation, and the data table. Add comments explaining any non-obvious layout decisions.

Multi-step research agent

Act as an autonomous research agent. Break the following task into parallel sub-tasks, assign each to a specialist role, and synthesise the results into a final report: compare the energy efficiency, cost trajectory, and adoption barriers of solid-state batteries versus lithium-iron-phosphate batteries for grid storage through 2030.

Hard math competition problem

Solve this problem step by step, showing all working and checking your answer: Find all pairs of positive integers (m, n) such that m squared minus n squared equals 2025, and prove that your list is complete.

Compare with

Kimi Latest
API-only model
1,048,576 token context · $0.29/M input, $24.00/M output
Kimi K3
Open-weights model
1,048,576 token context · $0.30/M input, $11.94/M output
Kimi K2 Thinking
Open-weights model
262,144 token context · $0.60/M input, $2.50/M output
Kimi K2.7 Code
Open-weights model
262,144 token context · $0.67/M input, $3.35/M output

More text models from moonshotai

Questions people ask

Are the weights open and can I run Kimi K2.5 locally?: Yes, the weights are published on Hugging Face at moonshotai/Kimi-K2.5. With 1T total parameters and 32B activated per token via MoE, you will need substantial GPU memory to run it locally — check the VRAM guides before committing.
What does Kimi K2.5 cost and what is its context window?: Via API it is $0.50 per million input tokens and $2.50 per million output tokens, with a 256K token context window.
How is Kimi K2.5 different from Kimi K2?: K2.5 is built on top of Kimi K2 Base with continued pretraining on approximately 15 trillion mixed visual and text tokens, adding native multimodality, visual coding capability, and the agent swarm execution paradigm that K2 lacked.

Captain's verdict
A rare open-weight model that takes vision, agentic orchestration, and hard reasoning seriously at once — worth the infrastructure cost if you need all three.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

AI Models for Image, Video & Text Generation

Kimi K2.5 is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models, check its heat score to see how it is trending, and open it on PirateDiffusion or BitVector to try it.

More Text / LLM models llm moonshotai vision reasoning agents coding long context

Browse by Model Family