Kimi K3: what the license allows, and what the maker charges
thecontinuitylayer.com/open-models/kimi-k3Moonshot AI's open-weight native multimodal agentic model: 2.8T total and 104B active parameters with a 1-million-token context window, served on the Kimi Open Platform as kimi-k3.
- 1Partly
Is Kimi K3 under an open license?
Partly, by Moonshot AI's own page.
Weights and code are under the Kimi K3 License, an MIT-style grant with conditions: a model-as-a-service business with over $20 million in revenue over 12 months needs a separate agreement with Moonshot AI, and a product with over 100 million monthly active users or $20 million monthly revenue must display "Kimi K3" on its interface.
Yes means: The weights are under an OSI-approved license such as Apache-2.0 or MIT. Partly means the maker's own license with conditions, such as a user cap or an acceptable-use clause; no means research or non-commercial use only.
- 2Yes
Can you download Kimi K3's weights?
Yes, by Moonshot AI's own page.
Weights are on Hugging Face under moonshotai, in MXFP4 from quantization-aware training.
Yes means: The maker's page links the weights for download with nothing more than a click-through license. Partly means an access form, an approval step, or only some sizes released.
- 3Yes
Does the maker state Kimi K3's context window?
Yes, by Moonshot AI's own page.
The card states a 1-million-token context window and lists a context length of 1048576.
Yes means: A token count for the context window is on the maker's page or model card.
- 4Partly
Does the maker say what it takes to run Kimi K3 on your own machine?
Partly, by Moonshot AI's own page.
2.8T total and 104B active parameters in MXFP4 weights; the card gives no GPU memory figure and points to vLLM, SGLang and TokenSpeed recipes and to community quantizations for llama.cpp, Ollama and LM Studio.
Yes means: The page states the parameter count and the memory or hardware it needs. Partly means the parameter count alone.
- 5Yes
Does the maker publish a price for its hosted Kimi K3?
Yes, by Moonshot AI's own page.
Kimi's pricing article lists the kimi-k3 API at $3.00 per million input tokens on a cache miss, $0.30 on a cache hit, and $15.00 per million output tokens. The platform pricing page's table did not render as text on the day checked.
Yes means: A per-token price on the maker's own pricing page. Partly means a hosted version exists but the price sits behind sign-in or a sales call.
- 6Yes
Does the maker train on your prompts when you use its hosted Kimi K3?
Yes, by Moonshot AI's own page.
The Kimi Open Platform model use agreement says Customer Content may be used to "provide, maintain, develop, support, and improve the Services" and that "Unless otherwise expressly agreed in writing, Customer Content may be used for the foregoing purposes"; a customer wanting restrictions on use for training must contact Moonshot AI.
Yes means: Here no is the good answer: the maker's page says what you send its API is not used for training. Yes means it is unless you opt out; partly means it depends on the plan or the page does not give the default.
- 7Partly
Can you fine-tune Kimi K3 and keep the result?
Partly, by Moonshot AI's own page.
The license grants the right to "run, deploy, fine-tune, or otherwise modify the Software and create derivative works" and to furnish copies with the notice kept, subject to the revenue and user thresholds above.
Yes means: The license allows fine-tuning and the fine-tuned weights are yours to keep and share. Partly means conditions follow the derivative, or a hosted fine-tune you cannot download.
License as stated: Kimi K3 License, stated on the model card for both the code and the weights.. Every grade above is read from Moonshot AI's own page on the day shown. If a page has changed, tell us.