Arnict

Arnict confirms zero data retention for inference traffic: prompts and outputs are not archived or used for training. Temporary inference caches are VRAM-only and expire 15 minutes after last use. Account, usage and billing metadata may be retained.

Models served by Arnict

Model name

Model	Context	Input /M	Output /M	Cache read /M	Cache hit	TTFT	Throughput
GLM 5.3 Flash Uncensored NVFP4 ZDR z-ai/glm-5.3-flash-uncensored:arnict	1.0M	$0.20/M	$0.799/M	$0.0699/M	91.9%	1.7s	142.8 tps
GLM 5.3 Uncensored NVFP4 ZDR z-ai/glm-5.3-uncensored:arnict	1.0M	$1.04/M	$2.09/M	$0.252/M	95.7%	3.3s	43.6 tps

GLM 5.3 Flash Uncensored

NVFP4

ZDR

z-ai/glm-5.3-flash-uncensored:arnict

Context

1.0M

Input /M

$0.20/M

Output /M

$0.799/M

Cache read /M

$0.0699/M

Cache hit

91.9%

TTFT

1.7s

Throughput

142.8 tps

GLM 5.3 Uncensored

NVFP4

ZDR

z-ai/glm-5.3-uncensored:arnict

Context

1.0M

Input /M

$1.04/M

Output /M

$2.09/M

Cache read /M

$0.252/M

Cache hit

95.7%

TTFT

3.3s

Throughput

43.6 tps

Prices are per million tokens. TTFT is the observed time to first token. 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 Arnict via the API

Call a model served by Arnict using its displayed model ID, including the provider suffix when selection is supported.

curl https://api.nano-gpt.com/api/v1/chat/completions \
  -H "Authorization: Bearer $NANOGPT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "z-ai/glm-5.3-flash-uncensored:arnict",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

For models with provider selection, append :arnict to the model ID, set "provider": "arnict" in the request body, or send an X-Provider: arnict 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.

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