Abliteration.ai

Data handling can vary by model and deployment. The privacy badge above shows the more conservative of the provider-wide fallback and the policies for routes on this page; each model page shows the policy for that specific route.

Abliteration.ai states that API prompts and completions are processed transiently, are not stored, and are not used for training.

Models served by Abliteration.ai

Model	Context	Input /M	Output /M	Cache read /M	Cache hit	Latency	Throughput
Abliterated Model BF16 abliteration-ai/abliterated-model:abliteration	262k	$1.05/M	$3.15/M	$0.11/M	0%	2.8s	62.3 tps
Abliterated Model Large FP8 abliteration-ai/abliterated-model-large:abliteration	1M	$3.15/M	$5.25/M	$0.31/M	20.7%	1.9s	113.2 tps
Abliterated Model Large V2 FP8 abliteration-ai/abliterated-model-large-v2:abliteration	1M	$3.15/M	$5.25/M	$0.31/M	78%	3.7s	124.7 tps

Prices are per million tokens. Selectable routes include the applicable provider-selection markup; fixed routes show the standard model price. Availability and pricing refresh continuously; each model page shows the live provider comparison.

Using Abliteration.ai via the API

For models with provider selection, append :abliteration to the model ID, set "provider": "abliteration" in the request body, or send an X-Provider: abliteration header. Explicit provider selection adds a route-specific markup over that provider's base price; the prices in the table above already include it. Models without provider selection use the displayed model ID without a provider suffix or selection surcharge.

curl https://nano-gpt.com/api/v1/chat/completions \
  -H "Authorization: Bearer $NANOGPT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "abliteration-ai/abliterated-model:abliteration",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

See the API documentation for provider preferences, price-aware routing, and error behavior.