# Unrestrictable > Abliterated and uncensored models on one prepaid key, through the OpenAI SDK and clients you already use. Priced per token, no subscription. Base URL: https://api.unrestrictable.ai/v1. Every model answers on the OpenAI Chat Completions, OpenAI Responses and Anthropic Messages endpoints. Keys and prepaid balance: https://app.unrestrictable.ai/console. Prices below are per million tokens as of 2026-11-16; the live list is https://unrestrictable.ai/models. ## Pages - [Models](https://unrestrictable.ai/models): every uncensored chat model with its prices - [Image and video](https://unrestrictable.ai/media): uncensored media models - [Uncensored models by family](https://unrestrictable.ai/uncensored): DeepSeek, Qwen, Gemma, GLM and Llama builds with counts, kinds and lowest prices - [Abliterated models](https://unrestrictable.ai/abliterated): every abliterated build with its price and context window - [Pricing](https://unrestrictable.ai/pricing): per-token prices, prepaid from $10, and the monthly usage plans - [Billing](https://unrestrictable.ai/billing): prepaid balance, what happens at zero, free requests, auto top-up and what is kept - [Docs](https://unrestrictable.ai/docs): SDK, SillyTavern, Janitor, OpenCode and Cline setup - [Guides](https://unrestrictable.ai/guides): step-by-step client setup - [SDKs](https://unrestrictable.ai/guides/sdks): the OpenAI SDK in Python, TypeScript, curl and the Vercel AI SDK - [Claude Code](https://unrestrictable.ai/guides/claude-code): an uncensored build inside Claude Code through claude-code-router - [Cursor](https://unrestrictable.ai/guides/cursor): custom key and base URL in Cursor - [LibreChat](https://unrestrictable.ai/guides/librechat): one custom endpoint in librechat.yaml with the uncensored ids - [Hermes Agent](https://unrestrictable.ai/guides/hermes-agent): a providers entry in ~/.hermes/config.yaml, transport chat_completions - [OpenClaw](https://unrestrictable.ai/guides/openclaw): a models.providers block in openclaw.json with api openai-completions - [Codex CLI](https://unrestrictable.ai/guides/codex-cli): a model_providers table in config.toml on the Responses API - [MCP server](https://api.unrestrictable.ai/mcp): every model on the key as MCP tools (chat, compare, search, embed, rerank, transcribe, media) for Claude Code, Cursor and any MCP client - [MCP discovery](https://unrestrictable.ai/.well-known/mcp.json): the same server as a machine-readable descriptor, for MCP directories and AI agents - [Media API](https://unrestrictable.ai/guides/media): image and video generation, polling, prices and refunds - [Playground](https://unrestrictable.ai/playground): chat with any listed model in the browser, free turns then a key - [Roleplay chat](https://unrestrictable.ai/roleplay): the same chat with character cards; cards export as SillyTavern JSON - [Red teaming](https://unrestrictable.ai/red-teaming): the security builds and the agent tools' settings - [Uncensored auto router](https://unrestrictable.ai/auto-router): model auto/uncensored, auto/abliterated, auto/security or auto/roleplay picks the build per request; an auto router key routes any app - [Compare](https://unrestrictable.ai/compare): two builds side by side with one costed month - [Custom models](https://unrestrictable.ai/custom-models): your own model id (persona, system prompt, defaults, fallbacks) on any uncensored or abliterated base, no training or hosting, priced at the base model - [Agents](https://unrestrictable.ai/agents): hosted agents with webhook skills, memory and cron on any listed model - [Workflows](https://unrestrictable.ai/workflows): a JSON graph of model calls run by one request - [Features](https://unrestrictable.ai/features): everything beyond chat, with live tool fees and the auto router's tiers - [Agent tools](https://unrestrictable.ai/features/agent-tools): Web search, web fetch, remote MCP servers and a code sandbox, run inside one request on an uncensored model. - [Files and redaction](https://unrestrictable.ai/features/files-and-redaction): PDFs and documents read by any model, and secrets or personal data swapped for placeholders before the request leaves. - [Auto router, fallbacks, presets](https://unrestrictable.ai/features/routing): model: "auto" picks from the uncensored tiers, a fallback list takes over on failure, a preset carries your system prompt. - [Response cache and long chats](https://unrestrictable.ai/features/caching): Exact repeats answered from the cache for a tenth of the charge; middle-out keeps a long roleplay inside the context window. - [Sessions and log export](https://unrestrictable.ai/features/observability): Every request of a chat or agent run on one timeline, CSV usage export, traces to your webhook, Langfuse or OTLP. - [Embeddings, audio, rerank, batch](https://unrestrictable.ai/features/endpoints): Embeddings, speech to text, reranking and an OpenAI-compatible Batch API on the same key. - [Price index](https://unrestrictable.ai/pricing-index): every listed model's prices as a table, JSON and CSV - [Tools](https://unrestrictable.ai/tools): every free tool on the site - [AI scenario generator](https://unrestrictable.ai/tools/ai-scenario-generator): roleplay scenarios written by an uncensored model, two free - [Status](https://unrestrictable.ai/status): the hourly API probe, the retry test and the models real traffic shows affected - [Verify a statement](https://unrestrictable.ai/statements/verify): check the signature on a monthly statement or a receipt - [Referral program](https://unrestrictable.ai/referrals): 10% of a referred account's top-ups for 12 months as credit, or 10% in USDT for 6 months - [Affiliate program](https://unrestrictable.ai/affiliates): the same program for guide writers, bot authors and community owners, paid as credit or USDT - [Reseller program](https://unrestrictable.ai/resellers): for operators who resell uncensored and roleplay models; keys per customer, written authorization, a rate set per partner - [Changelog](https://unrestrictable.ai/changelog): what changed, by day - [Brand](https://unrestrictable.ai/brand): the wordmark, palette and files - [Uncensored ChatGPT alternative](https://unrestrictable.ai/alternatives/chatgpt): ChatGPT's plans against a free chat and a per-token API that takes the OpenAI SDK - [OpenRouter alternative](https://unrestrictable.ai/alternatives/openrouter): fee, list and one costed month side by side - [kie.ai alternative](https://unrestrictable.ai/alternatives/kie-ai): uncensored image and video against a mainstream reseller - [Uncensy alternative](https://unrestrictable.ai/alternatives/uncensy): a chat app subscription against a per-token API behind your own front-end - [Pygmalion AI alternative](https://unrestrictable.ai/alternatives/pygmalion-ai): newer roleplay tunes on one key, with card import - [JannyAI](https://unrestrictable.ai/guides/jannyai): how to use the card directory, and the model to run its cards on - [KoboldAI Classic API](https://unrestrictable.ai/guides/koboldai-classic-api): the /api/v1/generate shape explained, and the OpenAI-format route to the same front-ends - [KoboldCpp remote API](https://unrestrictable.ai/guides/koboldcpp): KoboldCpp launched with --nomodel, Kobold Lite on the OpenAI Compatible provider, key and model id - [Uncensored coding AI](https://unrestrictable.ai/guides/uncensored-coding-ai): the code builds in Claude Code, Cursor and Open WebUI, a coding day costed live - [Ollama uncensored models](https://unrestrictable.ai/guides/ollama-uncensored-models): the Ollama library's uncensored tags next to the same builds by API, with prices - [Chub AI](https://unrestrictable.ai/guides/chub-ai): the reverse proxy URL and key in Secrets, the model id, starting settings and what 100 messages cost - [Wyvern Chat](https://unrestrictable.ai/guides/wyvern-chat-api): a Provider (API URL Type OpenAI, the URL, the stored key) and a Connection with the model id, on your own key - [LoreBary proxy](https://unrestrictable.ai/guides/lorebary-proxy): the three fields LoreBary's Proxy Portal passes to Janitor AI, Chub or Wyvern, what a custom proxy keeps, and the direct route - [Venice AI alternative](https://unrestrictable.ai/vs/venice): prepaid per token against a subscription - [Blog](https://unrestrictable.ai/blog): prices, leaderboards and setup notes, dated - [Proof](https://unrestrictable.ai/proof): the daily real-model check ## Data files and feeds - [Price index, JSON](https://unrestrictable.ai/pricing-index.json): every chat model's input, cached input and output price per million tokens and context window, with the catalog date - [Price index, CSV](https://unrestrictable.ai/pricing-index.csv): the same rows as CSV - [Daily check, JSON](https://unrestrictable.ai/proof.json): each model's verdict and dates - [Daily check, CSV](https://unrestrictable.ai/proof.csv): the same rows as CSV - [Blog feed](https://unrestrictable.ai/blog/feed.xml): RSS of published posts - [Changelog feed](https://unrestrictable.ai/changelog/feed.xml): RSS of the changelog - [API catalog](https://unrestrictable.ai/.well-known/api-catalog): RFC 9727 linkset - Markdown: the pages above plus .md (https://unrestrictable.ai/models.md, https://unrestrictable.ai/models/.md, https://unrestrictable.ai/index.md for the home page), or the page itself with Accept: text/markdown ## Usage plans A usage plan is a month of usage on every model. Bought by card, it renews automatically: the plan is bought again at its price when its usage balance falls below $5 or its month runs out, until you switch renewal off in Billing. Bought with crypto, it never renews. - Plus: $10 buys $20 of usage for a month - Pro: $25 buys $50 of usage for a month - Max: $100 buys $200 of usage for a month - Ultra: $250 buys $500 of usage for a month ## Chat models (170) - [GLM-5.3 Uncensored](https://unrestrictable.ai/models/glm-5.3-uncensored): glm-5.3-uncensored, uncensored, $0.282 in, $0.846 out, 99.8% success over the last 24 hours - [GLM-5.3 Abliterated](https://unrestrictable.ai/models/glm-5.3-abliterated): glm-5.3-abliterated, abliterated, $2.1825 in, $4.365 out, 99.9% success over the last 24 hours - [DeepSeek V4 Pro Uncensored](https://unrestrictable.ai/models/deepseek-v4-pro-uncensored): deepseek-v4-pro-uncensored, uncensored, $0.97 in, $1.94 out, 99.9% success over the last 24 hours - [GLM-5.3 Flash Abliterated](https://unrestrictable.ai/models/glm-5.3-flash-abliterated): glm-5.3-flash-abliterated, abliterated, $0.4656 in, $1.94 out, 100% success over the last 24 hours - [GLM-5.3 Flash Uncensored](https://unrestrictable.ai/models/glm-5.3-flash-uncensored): glm-5.3-flash-uncensored, uncensored, $0.1053 in, $0.3685 out, 100% success over the last 24 hours - [Kimi K3.1 Pro Uncensored](https://unrestrictable.ai/models/kimi-k3.1-pro-uncensored): kimi-k3.1-pro-uncensored, uncensored, $1.164 in, $5.82 out, 1048576 context, 98.5% success over the last 24 hours - [Kimi K3 Uncensored](https://unrestrictable.ai/models/kimi-k3-uncensored): kimi-k3-uncensored, uncensored, $5.745 in, $28.68 out, 262144 context, 90.7% success over the last 24 hours - [Nemotron 3 Ultra 550B Uncensored](https://unrestrictable.ai/models/nemotron-3-ultra-550b-a55b-uncensored): nemotron-3-ultra-550b-a55b-uncensored, uncensored, $1.0913 in, $6.402 out, 262144 context, 99.4% success over the last 24 hours - [DeepSeek V4.1 Flash Abliterated](https://unrestrictable.ai/models/deepseek-v4.1-flash-abliterated): deepseek-v4.1-flash-abliterated, abliterated, $0.8488 in, $1.6975 out, 100% success over the last 24 hours - [Dolphin Mistral 24B Venice Edition](https://unrestrictable.ai/models/dolphin-mistral-24b-venice): dolphin-mistral-24b-venice, uncensored, $0.50 in, $2.25 out, 32768 context - [Abliterated Large Flash](https://unrestrictable.ai/models/abliterated-model-large-flash): abliterated-model-large-flash, abliterated, $2.50 in, $5.00 out, 1048576 context, 100% success over the last 24 hours - [Abliterated Model](https://unrestrictable.ai/models/abliterated-model): abliterated-model, abliterated, $2.50 in, $7.50 out, 262144 context - [Abliterated Model Large](https://unrestrictable.ai/models/abliterated-model-large): abliterated-model-large, abliterated, $7.50 in, $12.50 out, 1000000 context - [Abliterated Model Large V2](https://unrestrictable.ai/models/abliterated-model-large-v2): abliterated-model-large-v2, abliterated, $7.4996 in, $12.4997 out, 960000 context - [Aion 3.5 Mini Uncensored](https://unrestrictable.ai/models/aion-3.5-mini-uncensored): aion-3.5-mini-uncensored, uncensored, per request, 120000 context - [Aion RP Llama 3.1 8B](https://unrestrictable.ai/models/aion-rp-llama-3.1-8b): aion-rp-llama-3.1-8b, roleplay, $3.20 in, $6.40 out, 32768 context - [Anubis 70B v1](https://unrestrictable.ai/models/anubis-70b-v1): anubis-70b-v1, roleplay, $1.24 in, $1.24 out, 65536 context - [Anubis 70B v1.1](https://unrestrictable.ai/models/anubis-70b-v1.1): anubis-70b-v1.1, roleplay, $1.085 in, $1.085 out, 32000 context - [Apex Flash 1 Abliterated](https://unrestrictable.ai/models/apex-flash-1-abliterated): apex-flash-1-abliterated, abliterated, $0.68 in, $2.32 out, 262144 context - [Artemis v1.1](https://unrestrictable.ai/models/artemis-v1.1): artemis-v1.1, roleplay, $0.35 in, $1.575 out, 262144 context - 150 more at https://unrestrictable.ai/llms-full.txt ## Image and video models (13) - [LTX-2.3 Image-to-Video Uncensored](https://unrestrictable.ai/media/ltx-2.3-i2v-uncensored): ltx-2.3-i2v-uncensored, video, from $0.08 per second - [Wan 3.0 Image-to-Video Uncensored](https://unrestrictable.ai/media/wan-3.0-i2v-uncensored): wan-3.0-i2v-uncensored, video, from $0.2 per second - [Wan 3.0 Video Uncensored](https://unrestrictable.ai/media/wan-3.0-video-uncensored): wan-3.0-video-uncensored, video, from $0.2 per second - [Anime Image Uncensored](https://unrestrictable.ai/media/anime-image-uncensored): anime-image-uncensored, image, from $0.03 per image - [Chroma Uncensored](https://unrestrictable.ai/media/chroma-uncensored): chroma-uncensored, image, from $0.08 per image - [CyberRealistic XL Uncensored](https://unrestrictable.ai/media/cyberrealistic-xl-uncensored): cyberrealistic-xl-uncensored, image, from $0.0136 per image - [Qwen Edit Uncensored](https://unrestrictable.ai/media/qwen-edit-uncensored): qwen-edit-uncensored, image, from $0.08 per image - [Qwen Image 2.1 Uncensored](https://unrestrictable.ai/media/qwen-image-2.1-uncensored): qwen-image-2.1-uncensored, image, from $0.0375 per image - [Qwen Image 3 Uncensored](https://unrestrictable.ai/media/qwen-image-3-uncensored): qwen-image-3-uncensored, image, from $0.08 per image - [Seedream 5.0 Lite Edit Uncensored](https://unrestrictable.ai/media/seedream-5.0-lite-edit-uncensored): seedream-5.0-lite-edit-uncensored, image, from $0.0313 per image - [Vision Image Uncensored](https://unrestrictable.ai/media/vision-image-uncensored): vision-image-uncensored, image, from $0.04 per image - [Vision Pro Image Uncensored](https://unrestrictable.ai/media/vision-pro-image-uncensored): vision-pro-image-uncensored, image, from $0.09 per image - [WAI Illustrious SDXL Uncensored](https://unrestrictable.ai/media/illustrious-xl-uncensored): illustrious-xl-uncensored, image, from $0.0136 per image