DeepSeek-V4.1-Flash: what the license allows, and what the maker charges
thecontinuitylayer.com/open-models/deepseek-v4-1-flashDeepSeek's current open release: a multimodal mixture-of-experts model with 552B backbone parameters, served on DeepSeek's API as deepseek-flash (version DeepSeek-V4.1-Flash-1017).
- 1Yes
Is DeepSeek-V4.1-Flash under an open license?
Yes, by DeepSeek's own page.
The card says the repository and the model weights are licensed under the MIT License, with the license file linked.
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 DeepSeek-V4.1-Flash's weights?
Yes, by DeepSeek's own page.
Weights are on Hugging Face under deepseek-ai, with 8-bit and FP8 tags and an inference folder for local setup; DeepSeek-V4.1-Flash-1017 is a second repository.
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 DeepSeek-V4.1-Flash's context window?
Yes, by DeepSeek's own page.
The card states support for contexts of up to one million tokens and lists a 1M-token context window in its sampling settings.
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 DeepSeek-V4.1-Flash on your own machine?
Partly, by DeepSeek's own page.
552B backbone parameters, with 8B active per token during prefill and 16B during decode; the card gives no GPU memory figure and points to the inference folder 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 DeepSeek-V4.1-Flash?
Yes, by DeepSeek's own page.
DeepSeek's pricing page lists deepseek-flash (DeepSeek-V4.1-Flash-1017) at $0.30 per million input tokens on a cache miss and $1.20 per million output tokens at peak, half that off-peak, and $0.006 per million on a cache hit, with a 1M context and 384K max output.
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.
- 6Not stated
Does the maker train on your prompts when you use its hosted DeepSeek-V4.1-Flash?
DeepSeek's pages do not say.
DeepSeek's Open Platform Terms of Service (effective 29 April 2026) do not say whether API inputs or outputs train DeepSeek's models and describe no opt-out; they defer personal information to the Privacy Policy, which says data is used to "operate, provide, develop, and improve the Services."
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.
- 7Yes
Can you fine-tune DeepSeek-V4.1-Flash and keep the result?
Yes, by DeepSeek's own page.
The card places the repository and weights under the MIT License and states no condition on fine-tuning, derivatives or redistribution.
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: MIT License, stated on the model card for the repository and the weights.. Every grade above is read from DeepSeek's own page on the day shown. If a page has changed, tell us.