# abliteratly

> OpenAI-compatible chat completions API for 8 uncensored models and 178 frontier models. Prepaid crypto billing, no email, no subscription, credits never expire. Only POST /v1/chat/completions and GET /v1/models are offered; there are no image, embedding, audio or responses endpoints.

## Start here
- [Uncensored AI API](https://abliteratly.beer/uncensored-ai-api): endpoint, models, limits and pricing.
- [Pricing](https://abliteratly.beer/pricing): live rates, request calculator, minimum $20 top-up.
- [Model directory](https://abliteratly.beer/models): every enabled model with context window and per-1M-token price.
- [Quickstart](https://abliteratly.beer/docs/quickstart): first request with curl, Python or Node.
- [OpenAI SDK](https://abliteratly.beer/docs/openai-sdk): SDK setup and unsupported endpoints.
- [OpenCode setup](https://abliteratly.beer/docs): one-command client installer.

## Guides and comparisons
- [Abliterated models](https://abliteratly.beer/guides/abliterated-models): how refusal-direction removal works and what it changes.
- [OpenRouter alternatives](https://abliteratly.beer/alternatives/openrouter): breadth vs an uncensored-first catalog.
- [Venice AI alternative](https://abliteratly.beer/alternatives/venice-ai): chat-first vs API-first.
- [Coding agents](https://abliteratly.beer/use-cases/coding-agents): tool loops, context and cost control.
- [Roleplay and fiction](https://abliteratly.beer/use-cases/roleplay): voice, memory and boundaries.
- [Methodology](https://abliteratly.beer/methodology): catalog, pricing and benchmark rules.
- [About](https://abliteratly.beer/about): what the service does and does not promise.

## Uncensored models
- [Magnum V4 72B](https://abliteratly.beer/models/magnum-v4-72b): Anthracite; 32.8K context; $12 input / $20 output per 1M tokens.
- [DeepSeek V4.1 Flash 1017 Uncensored](https://abliteratly.beer/models/deepseek-v4.1-flash-uncensored): DeepSeek; 1.05M context; $0.6 input / $2.4 output per 1M tokens.
- [Weaver (Alpha)](https://abliteratly.beer/models/weaver): Mancer; 8.0K context; $2 input / $3 output per 1M tokens.
- [Cydonia 24B V4.1](https://abliteratly.beer/models/cydonia-24b-v4.1): TheDrummer; 131.1K context; $1.2 input / $2 output per 1M tokens.
- [Skyfall 36B V2](https://abliteratly.beer/models/skyfall-36b-v2): TheDrummer; 32.8K context; $2.2 input / $3.2 output per 1M tokens.
- [UnslopNemo 12B](https://abliteratly.beer/models/unslopnemo-12b): TheDrummer; 1.02M context; $1.6 input / $1.6 output per 1M tokens.
- [ReMM SLERP 13B](https://abliteratly.beer/models/remm-slerp-l2-13b): Undi95; 6.1K context; $1.8 input / $2.6 output per 1M tokens.
- [GLM-5.3 Flash Uncensored](https://abliteratly.beer/models/glm-5.3-flash-uncensored): Z.ai; 1.05M context; $0.165 input / $0.618 output per 1M tokens.

## Policies
- [Terms](https://abliteratly.beer/terms)
- [Privacy](https://abliteratly.beer/privacy): prompt and response bodies are not stored by us; upstream retention varies by model.
- [Acceptable use](https://abliteratly.beer/aup)
- [Status](https://abliteratly.beer/status): observed 24-hour success rate and latency, not an SLA.