Are the weights open and can I run Kimi K2 0905 locally?

Yes, the weights are published on Hugging Face under moonshotai/Kimi-K2-Instruct-0905. With 1 trillion total parameters you will need substantial hardware to run it at full precision; most people will use a quantized version or the API.

What does Kimi K2 0905 cost and what is the context window?

Via API it is $0.60 per million input tokens and $2.50 per million output tokens, with a 256k-token context window.

How does the 0905 update differ from the original Kimi K2?

The September update improves agentic coding accuracy and generalisation across different scaffolds, strengthens frontend coding output, and extends the context window from 128k to 256k tokens.

Kimi K2 0905

1748°

#llm #moonshotai #coding #agents #reasoning #long context

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

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

MoonshotAI | 262,144 token context | $0.60/M input, $2.50/M output. Kimi K2 0905 is the September update of Kimi K2 0711. It is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It supports long-context inference up to 256k tokens, extended from the previous 128k. This update improves agentic coding with higher accuracy and better generalization across scaffolds, and enhances frontend coding with more aesthetic and functional outputs for web, 3D, and related tasks. Kimi K2 is optimized…MoonshotAI | 262,144 token context | $0.60/M input, $2.50/M output. Kimi K2 0905 is the September update of Kimi K2 0711. It is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It supports long-context inference up to 256k tokens, extended from the previous 128k. This update improves agentic coding with higher accuracy and better generalization across scaffolds, and enhances frontend coding with more aesthetic and functional outputs for web, 3D, and related tasks. Kimi K2 is optimized for agentic capabilities, including advanced tool use, reasoning, and code synthesis. It excels across coding (LiveCodeBench, SWE-bench), reasoning (ZebraLogic, GPQA), and tool-use (Tau2, AceBench) benchmarks. The model is trained with a novel stack incorporating the MuonClip optimizer for stable large-scale MoE training.

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

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Kimi K2 0905 is the September 2025 update of Moonshot AI's trillion-parameter MoE model, with 32 billion parameters active per forward pass and a 256k-token context window. The update sharpens agentic coding and frontend work, and the weights are open. It suits developers who want a serious coding agent they can self-host or run cheaply via API.

Known for

Strong agentic coding: competitive on SWE-Bench verified and SWE-Bench Multilingual.
Open weights on Hugging Face — you can download and run it yourself.
256k context window, doubled from the previous 128k release.
Improved frontend coding: better aesthetics and function for web and 3D tasks.
Tool use and reasoning built in, not bolted on.

Watch out for

1 trillion total parameters means local deployment demands serious hardware; most users will hit the API.
Output is priced at $2.50/M tokens — not expensive, but longer agentic runs add up fast.
Moonshot AI has not published the licence terms prominently; check the repo before building a product on it.

Best for

Prompts to try

Fix a real bug

Here is a Python function that is supposed to parse a nested JSON config and return a flat dict of dot-separated keys and their values. It silently drops keys when a value is an empty list. Identify the bug, explain why it happens, and return a corrected version with a test that would have caught it. ```python def flatten(d, prefix=''): out = {} for k, v in d.items(): key = f'{prefix}.{k}' if prefix else k if isinstance(v, dict): out.update(flatten(v, key)) elif v: out[key] = v return out ```

Build a small web component

Write a self-contained HTML file with no external dependencies that renders an interactive kanban board with three columns: To Do, In Progress, Done. Cards should be draggable between columns using the HTML drag-and-drop API. Style it cleanly with inline CSS. Add a small form at the top to create a new card with a title.

Plan a multi-step agent task

You are a coding agent with access to a bash terminal and a file editor. A user's Node.js project fails its CI pipeline because three tests import a module that no longer exists after a recent refactor. Describe the exact sequence of tool calls you would make — what you read first, how you trace the broken imports, and how you decide what to write — before touching any file.

Compare with

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
Kimi K2.5
Open-weights model
262,144 token context · $0.50/M input, $2.50/M output

More text models from moonshotai

Questions people ask

Are the weights open and can I run Kimi K2 0905 locally?: Yes, the weights are published on Hugging Face under moonshotai/Kimi-K2-Instruct-0905. With 1 trillion total parameters you will need substantial hardware to run it at full precision; most people will use a quantized version or the API.
What does Kimi K2 0905 cost and what is the context window?: Via API it is $0.60 per million input tokens and $2.50 per million output tokens, with a 256k-token context window.
How does the 0905 update differ from the original Kimi K2?: The September update improves agentic coding accuracy and generalisation across different scaffolds, strengthens frontend coding output, and extends the context window from 128k to 256k tokens.

Captain's verdict
A capable open-weights coding agent at a reasonable API price — the 256k context and agentic improvements make the September update worth the switch.

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.

Similar Models

GPT 3.5 Turbo Instruct

TEXT / LLM

Mistral Medium 3.1

TEXT / LLM

AI Models for Image, Video & Text Generation

Kimi K2 0905 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 and check its heat score to see how it is trending.

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

Browse by Model Family