RunInfra

RunInfra states that prompts and completions are not used to train or improve models. Prompt prefixes may remain in a serving-machine cache, and non-streaming responses requested with an Idempotency-Key may be retained for up to 24 hours for replay. Usage and billing metadata are retained. This route is not classified as Zero Data Retention.

Models served by RunInfra

Model name

Model	Context	Input /M	Output /M	Cache read /M	TTFT	Throughput
GLM 5.3 NVFP4 Not ZDR z-ai/glm-5.3:runinfra	1.0M	$1.36/M	$4.41/M	$0.136/M	N/A	N/A
GLM 5.3 (Thinking) NVFP4 Not ZDR z-ai/glm-5.3:thinking:runinfra	1.0M	$1.36/M	$4.41/M	$0.136/M	N/A	N/A
GLM 5.3 Flash Not ZDR z-ai/glm-5.3-flash:runinfra	1.0M	$0.116/M	$0.472/M	$0.0315/M	N/A	N/A

GLM 5.3

NVFP4

Not ZDR

z-ai/glm-5.3:runinfra

Context

1.0M

Input /M

$1.36/M

Output /M

$4.41/M

Cache read /M

$0.136/M

TTFT

N/A

Throughput

N/A

GLM 5.3 (Thinking)

NVFP4

Not ZDR

z-ai/glm-5.3:thinking:runinfra

Context

1.0M

Input /M

$1.36/M

Output /M

$4.41/M

Cache read /M

$0.136/M

TTFT

N/A

Throughput

N/A

GLM 5.3 Flash

Not ZDR

z-ai/glm-5.3-flash:runinfra

Context

1.0M

Input /M

$0.116/M

Output /M

$0.472/M

Cache read /M

$0.0315/M

TTFT

N/A

Throughput

N/A

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 RunInfra via the API

Call a model served by RunInfra 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:runinfra",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

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