Modelle/Character Chat LLM API

Aktiv12 models

Character Chat LLM API — Rollenspiel- und Dialogmodelle mit Persönlichkeit

Spezialisierte Sprachmodelle für immersives Rollenspiel, interaktives Storytelling und lebensechte Charakterkonversationen auf MuAPI.Unterstützt Cydonia, Anubis, Euryale, MythoMax und weitere Modelle für dynamische virtuelle Welten und kreative Fiktion.

Get API Key Try Cydonia 24B v4.1 in Playground

Cydonia & Aion

Cydonia & Aion models2 models

Text GenerationLiveFeatured

Cydonia & Aion

Cydonia 24B v4.1

Cydonia 24B v4.1, a community fine-tune tuned for creative writing, character chat, and interactive fiction, with a 131K-token context window.

Text in

$0.70/M in · $1.10/M out

Modell testen

Text GenerationLive

Cydonia & Aion

Aion Llama 3.1 8B

Aion Llama 3.1 8B, a compact fine-tune of Llama 3.1 8B for character chat with a 32K-token context window.

Text in

$1.60/M in · $3.20/M out

Modell testen

Llama-based character chat

Llama-based character chat models8 models

Text GenerationLive

Llama-based character chat

Anubis 70B v1.1

Anubis 70B v1.1, a community Llama 70B fine-tune tuned for creative writing and character chat, with a 32K-token context window.

Text in

$0.62/M in · $0.62/M out

Modell testen

Text GenerationLive

Llama-based character chat

Euryale 70B v2.3

Euryale 70B v2.3, a community Llama 3.3 70B fine-tune tuned for creative writing and character chat, with a 20K-token context window.

Text in

$0.986/M in · $0.986/M out

Modell testen

Text GenerationLive

Llama-based character chat

Stheno 8B v3.2

Stheno 8B v3.2, a community Llama 3 8B fine-tune tuned for character chat and creative writing, with a 16K-token context window.

Text in

$0.401/M in · $0.401/M out

Modell testen

Text GenerationLive

Llama-based character chat

EVA Llama 3.33 70B

EVA Llama 3.33 70B, a community fine-tune tuned for story writing and character chat, with a 32K-token context window.

Text in

$4.012/M in · $4.012/M out

Modell testen

Text GenerationLive

Llama-based character chat

MythoMax 13B

MythoMax L2 13B, a widely used community merge for storytelling and character chat, with a 4K-token context window.

Text in

$0.201/M in · $0.201/M out

Modell testen

Text GenerationLive

Llama-based character chat

Wayfarer 70B

Wayfarer Large 70B, a Llama 3.3 fine-tune tuned for interactive fiction and text-adventure style character chat, with a 32K-token context window.

Text in

$1.40/M in · $1.40/M out

Modell testen

Text GenerationLive

Llama-based character chat

Nevoria R1 70B

Nevoria R1 70B, a community Llama 3.3 merge tuned for creative writing and character chat, with a 32K-token context window.

Text in

$0.986/M in · $0.986/M out

Modell testen

Text GenerationLive

Llama-based character chat

Magnum v4 72B

Magnum v4 72B, a community Qwen 72B fine-tune tuned for prose quality in creative writing and character chat, with a 16K-token context window.

Text in

$4.012/M in · $5.984/M out

Modell testen

Qwen character chat

Qwen character chat models2 models

Text GenerationLive

Qwen character chat

Qwen 3.8 27B Queen

Qwen 3.8 27B Queen, a community fine-tune tuned for character chat and creative chat, with image input and a 524K-token context window.

Text + Image in

$0.50/M in · $3.00/M out

Modell testen

Text GenerationLive

Qwen character chat

Qwen 3.8 27B Fable

Qwen 3.8 27B Fable, a community fine-tune tuned for long-form storytelling, with image input and a 524K-token context window.

Text + Image in

$0.50/M in · $3.00/M out

Modell testen

What is the Character Chat LLM API?

The Character Chat LLM API is Muapi's catalog of 12 character chat and creative-writing LLMs, from lightweight 8B chat models to 1M-token-context reasoning models, behind one consistent REST interface so you can compare them by changing a single model name.

Senden Sie einen Prompt, optional mit einem System-Prompt zur Definition von Persönlichkeit und Tonfall sowie einer Bild-URL bei vision-fähigen Modellen, und rufen Sie den generierten Text ab. Typische Anwendungsfälle umfassen Charakter-Rollenspiele, interaktive Geschichten, ausführliches Storytelling und eigene Chat-Anwendungen. Dies sind feinabgestimmte Community-Modelle für kreatives Schreiben und Dialoge – sie sind nicht abliterated. Die Nutzungsrichtlinien von MuAPI gelten für alle generierten Inhalte. Sie bleiben für die Verwendung der Ausgaben verantwortlich.

Hauptfunktionen

12 Models, One API

Every endpoint accepts the same request body and returns text through the shared result endpoint, so switching models is a one-line change.

Tuned for Creative Writing

These models are fine-tuned on fiction and character chat data, so they tend to hold a character, keep a narrative voice, and write longer, more vivid prose than general assistants.

System Prompt Control

Set a system_prompt to steer persona, tone, and output format for creative, analytical, or evaluation work.

Pay Per Token

Billing follows the token counts each call returns. There is no subscription and no model hosting to manage.

Image Input on Vision Models

Pass an image_url to Qwen 3.8 27B Queen, Qwen 3.8 27B Fable.

Acceptable-Use Policy Applies

Reduced refusals do not mean no rules. Muapi's acceptable-use policy and applicable law still apply to what you generate.

Pricing

Endpoint	Input	Output	Price	Best For
cydonia-24b-v4-1	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$0.70/M in · $1.10/M out	Long-context character chat and story writing at a low token rate
aion-llama-3-1-8b	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$1.60/M in · $3.20/M out	Compact character chat and chat
anubis-70b-v1-1	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$0.62/M in · $0.62/M out	Large-model character chat and storytelling at a flat, low rate
euryale-70b-v2-3	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$0.986/M in · $0.986/M out	Character chat and interactive fiction
stheno-8b-v3-2	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$0.401/M in · $0.401/M out	Fast, inexpensive character chat and chat
eva-llama-3-33-70b	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$4.012/M in · $4.012/M out	Long-form story writing and character chat
mythomax-13b	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$0.201/M in · $0.201/M out	The lowest-cost option for short storytelling and character chat
wayfarer-70b	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$1.40/M in · $1.40/M out	Text adventures and interactive fiction
nevoria-r1-70b	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$0.986/M in · $0.986/M out	Creative writing and character chat with a reasoning-tuned base
magnum-v4-72b	Prompt + system_prompt, max_tokens	Complete text response (poll for result)	$4.012/M in · $5.984/M out	High-quality prose for creative writing and character chat
qwen-3-8-27b-queen	Prompt + system_prompt, max_tokens, optional image_url	Complete text response (poll for result)	$0.50/M in · $3.00/M out	Character chat with image input and very long context
qwen-3-8-27b-fable	Prompt + system_prompt, max_tokens, optional image_url	Complete text response (poll for result)	$0.50/M in · $3.00/M out	Long-form storytelling with image input and very long context

Billed per token from the usage each call returns; prices are per million input and output tokens.

How to Call the Character Chat LLM API

Submit a JSON request with your Muapi API key, save the request ID, and poll the shared result endpoint.

1. Prompt an Cydonia 24B v4.1 senden

curl -X POST https://api.muapi.ai/api/v1/cydonia-24b-v4-1 \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "You are Mira, a wry tavern keeper. Greet a stranger who just walked in from the rain.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call cydonia-24b-v4-1 with a prompt and optional system_prompt.

2. Prompt an Aion Llama 3.1 8B senden

curl -X POST https://api.muapi.ai/api/v1/aion-llama-3-1-8b \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Play a friendly dragon guarding a library and answer my first question.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call aion-llama-3-1-8b with a prompt and optional system_prompt.

3. Prompt an Anubis 70B v1.1 senden

curl -X POST https://api.muapi.ai/api/v1/anubis-70b-v1-1 \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Continue this story: the lighthouse keeper heard a knock at the door at midnight.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call anubis-70b-v1-1 with a prompt and optional system_prompt.

4. Prompt an Euryale 70B v2.3 senden

curl -X POST https://api.muapi.ai/api/v1/euryale-70b-v2-3 \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Act as a rival knight and challenge me to a duel with a short speech.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call euryale-70b-v2-3 with a prompt and optional system_prompt.

5. Prompt an Stheno 8B v3.2 senden

curl -X POST https://api.muapi.ai/api/v1/stheno-8b-v3-2 \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Play a cheerful robot sidekick and react to my plan.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call stheno-8b-v3-2 with a prompt and optional system_prompt.

6. Prompt an EVA Llama 3.33 70B senden

curl -X POST https://api.muapi.ai/api/v1/eva-llama-3-33-70b \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Write the opening chapter of a slow-burn mystery novel.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call eva-llama-3-33-70b with a prompt and optional system_prompt.

7. Prompt an MythoMax 13B senden

curl -X POST https://api.muapi.ai/api/v1/mythomax-13b \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Tell a short fairy tale about a fox who steals moonlight.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call mythomax-13b with a prompt and optional system_prompt.

8. Prompt an Wayfarer 70B senden

curl -X POST https://api.muapi.ai/api/v1/wayfarer-70b \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "You are the narrator of a text adventure. I wake up in a locked cellar. What do I see?", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call wayfarer-70b with a prompt and optional system_prompt.

9. Prompt an Nevoria R1 70B senden

curl -X POST https://api.muapi.ai/api/v1/nevoria-r1-70b \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Continue this scene, keeping the narrator voice and pacing.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call nevoria-r1-70b with a prompt and optional system_prompt.

10. Prompt an Magnum v4 72B senden

curl -X POST https://api.muapi.ai/api/v1/magnum-v4-72b \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Write a vivid, atmospheric paragraph describing a city at dawn after a storm.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call magnum-v4-72b with a prompt and optional system_prompt.

11. Prompt an Qwen 3.8 27B Queen senden

curl -X POST https://api.muapi.ai/api/v1/qwen-3-8-27b-queen \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "You are a regal elf queen. Receive the messenger and respond in character.", "system_prompt": "You are a candid, detail-oriented assistant.", "image_url": "https://example.com/reference.png", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call qwen-3-8-27b-queen with a prompt and optional system_prompt, image_url.

12. Prompt an Qwen 3.8 27B Fable senden

curl -X POST https://api.muapi.ai/api/v1/qwen-3-8-27b-fable \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Tell a long-form bedtime fable about a clockmaker and a wandering star.", "system_prompt": "You are a candid, detail-oriented assistant.", "image_url": "https://example.com/reference.png", "max_tokens": 1024}'

# Response: {"request_id":"REQUEST_ID"}

Call qwen-3-8-27b-fable with a prompt and optional system_prompt, image_url.

13. Ergebnis abrufen

curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result \
  -H "x-api-key: YOUR_API_KEY"

# Poll until status is completed, then read the generated text from outputs.

Poll the shared result endpoint until the task is completed, then read the generated text from outputs.

Häufig gestellte Fragen

What is a character chat LLM?

A character chat LLM is a language model fine-tuned on fiction, dialogue and character chat data so it holds a character, keeps a narrative voice, and writes longer, more vivid prose. The models on this page are community fine-tunes tuned for creative writing, not abliterated versions of general assistants.

Which models does the Character Chat LLM API include?

Cydonia 24B v4.1, Aion Llama 3.1 8B, Anubis 70B v1.1, Euryale 70B v2.3, Stheno 8B v3.2, EVA Llama 3.33 70B, MythoMax 13B, Wayfarer 70B, Nevoria R1 70B, Magnum v4 72B, Qwen 3.8 27B Queen and Qwen 3.8 27B Fable. Each is a separate endpoint with its own model name in the URL.

How much does it cost?

Billing is per token, deducted after each call from the counts the model returns. Rates start at $0.201 per million input tokens; see the pricing table for every model.

Will these models answer anything I ask?

No. Reduced refusals do not mean unrestricted output, and each model's knowledge and quality still bound the answer. You are responsible for how you use the output, and requests that violate Muapi's acceptable-use policy or the law are not permitted.

Can I send images to these models?

Yes on Qwen 3.8 27B Queen, Qwen 3.8 27B Fable: add an image_url next to your prompt. The others are text-only.

How do I get the result?

POST to /api/v1/<model-name> returns a request_id. Poll /api/v1/predictions/{request_id}/result until the status is completed, then read the text from outputs.

Can I use these for a character or storytelling app?

Yes: set a system_prompt with the character's persona and send the running conversation in the prompt. Muapi's acceptable-use policy applies, and you are responsible for moderating your own product.

Explore the Llama Character Chat API →8 models.Explore the Cydonia Character Chat API →Cydonia 24B v4.1 and Aion Llama 3.1 8B Explore the Qwen Character Chat API →Qwen 3.8 27B Queen and Qwen 3.8 27B Fable See the Abliterated LLM API →20 models on the Abliterated LLM API.See the Low-Refusal LLM API →2 models on the Low-Refusal LLM API.See Gemma 4 31B SDFT Abliterated →An abliterated Gemma 4 model tuned for character chat, listed under the Gemma abliterated hub.Browse every LLM on Muapi →Compare these models against GPT, Claude, Gemini, DeepSeek and every other language model in the catalog.

Ready to build with the Character Chat LLM API?

Create one Muapi API key and call every model on this page with a plain REST request.

Get an API Key Try Cydonia 24B v4.1 in Playground