UncensoredLabs

UncensoredLabs provides uncensored and abliterated AI models for developers building chat, coding and AI applications. The README describes access to uncensored and abliterated DeepSeek and GLM text models through one OpenAI-compatible API, or directly in browser Chat. The service is built in Switzerland and describes itself as independent, not affiliated with OpenAI, DeepSeek or Zhipu AI. The README defines abliterated as models whose weights have been modified to reduce refusal behavior, and adds that this does not promise a correct answer or no refusals on every prompt.

To connect, keep your existing OpenAI client, change the base URL and use an UncensoredLabs application key. Chat Completions supports regular responses and streaming. Billing is prepaid and token-based, with API-key spending limits, model allowlists and usage tracking. Eligible new accounts receive a one-time $0.10 trial credit, and prepaid top-ups start at $3. Normal token pricing and limits apply.

The documentation says the gateway checks the key, model permissions, limits and available prepaid credit before forwarding a request to the configured inference provider. Application servers do not store prompts or model responses, and provider retention policies apply. The README lists use cases such as security research, reverse engineering, developer tools, creative work and trust and safety. The public GitHub repository contains documentation, configuration and integration examples, not the hosted platform source.