Updated

Abliterated LLM API providers

Seven providers sell API access to abliterated or uncensored models, and the formats they accept decide which SDKs and agents work. Silk Compute ranks first, with all three formats and a 1M-token context on every model.

abliteratedllmapi.com is published by the team behind Silk Compute, one of the providers ranked here.

ProviderChat CompletionsMessagesResponsesContext1M-context pricePrompts storedCrypto
1Silk Compute1M on every model$0.24 / $0.90DeepSeek V4.1 Flash 1017Zero data retention
2Abliteration1M on Large v2$1.00 / $2.00Abliterated Large Flash"never stored by default"
3Morph1M$0.30 / $1.20DeepSeek V4.1 Flash UncensoredKept up to 30 days on free and paid tiers
4Unfil.AI128KNo 1M model"We do not store your API prompts or responses after processing"
5Featherless32KNo 1M model"does not log chats, prompts, or completions"
6Awan LLM8KNo 1M model"We do not log user prompts and generations."
7ModelsLabNot publishedNo 1M modelPrivacy policy says nothing about prompts

Prices are in / out per 1M tokens. Ranked by formats served, then prompt storage, context and price, from each provider's own pages read November 18, 2026.

Each provider

1

Silk Compute

Visit Silk Compute

https://api.silkcompute.com/chat/completions

Strengths

  • All three formats on every model.
  • Card, or Monero, Bitcoin, Ether, USDT and USDC.

Shortfalls

  • An account runs 8 requests at once until support raises it.
  • Hosted tools such as web search are not available.

Source silkcompute.com/llms.txt

2

Abliteration

Visit Abliteration

Chat Completions, Responses, Messages

Strengths

  • All three formats.
  • Guides for Claude Code and Codex.

Shortfalls

  • No crypto payment listed.
  • Its Flash model is $1.00 / $2.00 per 1M.

Source docs.abliteration.ai/models

4

Unfil.AI

Visit Unfil.AI

https://api.unfil.ai/v1/chat/completions

Strengths

  • Hermes 3 405B with 128K context.
  • Card, wallet or crypto.

Shortfalls

  • Its docs page says coming soon.
  • Usage logs with IP addresses kept 30 days.

Source unfil.ai/privacy-policy

5

Featherless

Visit Featherless

https://api.featherless.ai/v1/chat/completions

Strengths

  • Works with Cline, Kilo Code, OpenCode and OpenHands on its developer plan.
  • States it does not log prompts.

Shortfalls

  • 32K context on Qwen 3.8 27B Abliterated.
  • Card only.

Source featherless.ai/llms.txt

6

Awan LLM

Visit Awan LLM

https://api.awanllm.com/v1/chat/completions

Strengths

  • Chat and text completion endpoints.
  • States it does not log prompts.

Shortfalls

  • 8K context on Llama 3 8B Abliterated.
  • PayPal only.

Source www.awanllm.com/quick-start

7

ModelsLab

Visit ModelsLab

/api/v7/llm/chat/completions

Strengths

  • One OpenAI-compatible endpoint for every model.
  • uncensored-chat is $0.30 / $1.20 per 1M.

Shortfalls

  • No base model named for uncensored-chat.
  • A subscription from $21 a month is required.

Source modelslab.com/uncensored-ai-api