Updated
About abliteratedllmapi
abliteratedllmapi is a reference for calling abliterated LLMs over an API. abliteratedllmapi.com is published by the team behind Silk Compute, one of the providers ranked here.
Why Silk Compute is the worked example
It ranks first on the providers page, and it serves Chat Completions, Messages and Responses on every model, so one base URL runs every sample here. Silk Compute does not store your prompts or model outputs.
How providers are ranked
By request formats served, then whether prompts are stored, then context, then the price of a 1M-context model. Only facts from each provider's own pages count.
Sources
| Fact | Source |
|---|---|
| Endpoints, parameters, limits, errors | silkcompute.com/docs |
| Model ids, prices, max output | silkcompute.com/llms.txt |
| Other providers | Each provider's own pricing, docs and privacy pages, linked on the providers page |
| SDK usage | openai-go, openai-node, anthropic-sdk-typescript |
| Abliteration | Arditi et al., 2024 and the abliteration write-up on Hugging Face |
Everything was read November 18, 2026. Prices and policies change, so check the source before you commit.
abliteratedllmapi.com is published by the team behind Silk Compute, one of the providers ranked here.