Models·TheDrummer

Skyfall 36B V2: Creative Text Model Pricing

thedrummer/skyfall-36b-v2

An open-weight, text-only model tuned for creative writing, role-playing, and coherent stories.

$0.55Input, per million tokens
$0.80Output, per million tokens
33KContext window
29KLongest answer

33K tokens of context — about 24,576 words, on a scale that ends at 2M.

What it can do

  • Open weights
  • Takes text
  • Gives text

Released 10 Mar 2025 · checked 3 Oct 2026.

About it

Skyfall 36B V2 is TheDrummer’s text model, released March 10, 2025. It is an enhanced version of Mistral Small 2501, fine-tuned for creative and nuanced writing, role-playing, and connected storytelling. The model has open weights, a 32,768-token context window, and prices for both input and output tokens.

Review

In a line

Skyfall 36B V2 is a text-only language model from TheDrummer. It is an enhanced, fine-tuned version of Mistral Small 2501, released March 10, 2025.

What it's good at

The model is tuned for creative writing, nuanced wording, role-playing, and coherent storytelling. These uses suit people who want responses that maintain a clear thread across a longer piece of text.

It accepts text and returns text; image input and output are not included.

Price in plain terms

Billing is $0.55 per million input tokens and $0.8 per million output tokens. Tokens are small pieces of text counted for billing.

Reading a page of about 500 words, or roughly 650 input tokens, costs about 0.036 cents. A chat turn with about 1,000 input tokens and 500 output tokens costs about 0.095 cents. At that rate, 1,000 such turns cost about $0.95.

Context and limits

Skyfall 36B V2 can take up to 32,768 tokens in one conversation context. It can generate up to 29,491 tokens in one answer.

It works with text only, so the listed inputs and outputs do not include images or other media.

Ways to use it

The model is marked as open weights, so it can be used in setups that support open-weight models. Its Hugging Face identifier is TheDrummer/Skyfall-36B-v2, and its model ID is thedrummer/skyfall-36b-v2.

Verdict

Choose Skyfall 36B V2 if your main goal is text generation with a creative, role-playing, or storytelling focus and you want open weights. Its token prices are low enough for many short exchanges, with separate charges for what you send and what it writes.

Look elsewhere if you need image handling, because this model is text-only. Its 32,768-token context and 29,491-token output ceiling also set a fixed boundary for very large documents or responses.

Verdict

Skyfall 36B V2 fits writers, role-playing projects, and story-focused chat tools that can work with open-weight models. Its pricing is easy to estimate, and the model offers a large text context for extended exchanges. It is not a match for image-based tasks, since it accepts and returns text only. Check that your chosen setup supports the model’s open weights and stays within its context and response limits before committing to it.

  • Fine-tuned for creative and nuanced writing
  • Designed for role-playing and coherent storytelling
  • Open weights available
  • 32,768-token context window
  • Input and output prices are listed separately
  • Text only; no image handling is listed
  • Maximum response length is 29,491 tokens
  • Requires a setup that supports its open weights
  • Pick it ifCreative writing and story generation; Role-playing with text responses; Users comparing low per-token costs
  • Skip it ifYou need image input or image output; You need responses beyond 29,491 tokens; Your setup cannot use open-weight models

Pricing

Input is $0.55 per million tokens and output is $0.8 per million tokens. A 500-word page costs about 0.036 cents to read; 1,000 typical chat turns cost about $0.95.

Questions

What is Skyfall 36B V2?

It is TheDrummer’s text-only language model, released March 10, 2025, and fine-tuned from Mistral Small 2501.

What is Skyfall 36B V2 good for?

It is tuned for creative writing, nuanced writing, role-playing, and coherent storytelling.

How much does it cost?

Input costs $0.55 per million tokens and output costs $0.8 per million tokens.

How much does a typical chat turn cost?

A turn using about 1,000 input tokens and 500 output tokens costs about 0.095 cents. 1,000 such turns cost about $0.95.

How much text can it read at once?

Its context window is 32,768 tokens.

How long can one answer be?

The maximum output is 29,491 tokens.

Does it support images?

No. The listed input and output modality is text only.

Where is the model identified?

Its Hugging Face ID is TheDrummer/Skyfall-36B-v2, and its model ID is thedrummer/skyfall-36b-v2.

On these boards

More from TheDrummer