Meta Llama 3.3 70B und 8B mit entfernten Verweigerungsschichten für freie Textgestaltung, sachliche Recherchen und offene Inferenz auf MuAPI.
Llama 3.3 70B Instruct with refusal behavior reduced by abliteration: a large open-weights chat baseline with a 32K-token context window.
NeuralDaredevil 8B, an abliterated Llama 3 8B derivative: the lightest and cheapest option here, with an 8K-token context window.
This page covers Llama 3.3 70B Instruct Abliterated and NeuralDaredevil 8B Abliterated on Muapi. Llama 3.3 70B Instruct with refusal behavior reduced by abliteration: a large open-weights chat baseline with a 32K-token context window. NeuralDaredevil 8B, an abliterated Llama 3 8B derivative: the lightest and cheapest option here, with an 8K-token context window.
Senden Sie einen Prompt, optional mit einem System-Prompt zur Vorgabe von Rolle und Tonfall sowie einer Bild-URL bei vision-fähigen Modellen, und fragen Sie den generierten Text ab. Typische Einsatzbereiche sind kreatives Schreiben, Red-Teaming- und Sicherheitsforschung sowie Szenarien, in denen Standard-Verweigerungen legitime Arbeitsabläufe behindern. Abliterated Modelle sind Sprachmodelle mit reduzierter Verweigerungsneigung; sie garantieren keine vollkommen unbeschränkte Ausgabe und heben die Nutzungsrichtlinien von MuAPI nicht auf. Sie tragen die Verantwortung für die Verwendung der generierten Ergebnisse.
Every endpoint accepts the same request body and returns text through the shared result endpoint, so switching models is a one-line change.
Abliteration and related techniques lower how often the model declines a request. Behavior varies by model and prompt, and is not a guarantee.
Set a system_prompt to steer persona, tone, and output format for creative, analytical, or evaluation work.
Billing follows the token counts each call returns. There is no subscription and no model hosting to manage.
Use these models for chat, drafting, and step-by-step reasoning.
Reduced refusals do not mean no rules. Muapi's acceptable-use policy and applicable law still apply to what you generate.
| Endpoint | Input | Output | Price | Best For |
|---|---|---|---|---|
| llama-3-3-70b-abliterated | Prompt + system_prompt, max_tokens | Complete text response (poll for result) | $1.40/M in · $1.40/M out | Large open-weights chat at a flat token rate |
| neuraldaredevil-8b-abliterated | Prompt + system_prompt, max_tokens | Complete text response (poll for result) | $0.88/M in · $0.88/M out | Lightweight, low-cost chat and prototyping |
Billed per token from the usage each call returns; prices are per million input and output tokens.
Submit a JSON request with your Muapi API key, save the request ID, and poll the shared result endpoint.
curl -X POST https://api.muapi.ai/api/v1/llama-3-3-70b-abliterated \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt": "Continue this interactive fiction scene without softening the tone.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'
# Response: {"request_id":"REQUEST_ID"}Call llama-3-3-70b-abliterated with a prompt and optional system_prompt.
curl -X POST https://api.muapi.ai/api/v1/neuraldaredevil-8b-abliterated \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt": "Give me five brainstorming angles for a horror short story.", "system_prompt": "You are a candid, detail-oriented assistant.", "max_tokens": 1024}'
# Response: {"request_id":"REQUEST_ID"}Call neuraldaredevil-8b-abliterated with a prompt and optional system_prompt.
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.
An abliterated language model is a model whose refusal behavior has been reduced by editing the internal direction that drives refusals, usually without retraining from scratch. Related terms such as obliterated and derestricted describe similar techniques. The result tends to decline fewer requests than the original model, but it is not a guarantee and quality can differ from the base model.
Llama 3.3 70B Instruct Abliterated and NeuralDaredevil 8B Abliterated. Each is a separate endpoint with its own model name in the URL.
Billing is per token, deducted after each call from the counts the model returns. Rates start at $0.88 per million input tokens; see the pricing table for every model.
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.
No. Every model on this page is text in, text out.
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.
Create one Muapi API key and call every model on this page with a plain REST request.