Lilac

Models served by Lilac

Model	Context	Input /M	Output /M	Cache read /M	Cache hit	Latency	Throughput
Gemma 4 31B google/gemma-4-31b-it:lilac	262k	$0.12/M	$0.37/M	N/A	35.3%	1.8s	46 tps
Gemma 4 31B (Thinking) google/gemma-4-31b-it:thinking:lilac	262k	$0.12/M	$0.37/M	N/A	35.3%	1.8s	46 tps
GLM 5.2 FP8 z-ai/glm-5.2:lilac	1.0M	$0.94/M	$3.15/M	$0.18/M	80.6%	2.1s	109.4 tps
GLM 5.2 (Thinking) FP8 z-ai/glm-5.2:thinking:lilac	1.0M	$0.94/M	$3.15/M	$0.18/M	78.9%	2.1s	109.4 tps
Kimi K2.6 INT4 moonshotai/kimi-k2.6:lilac	256k	$0.73/M	$3.68/M	$0.17/M	N/A	3.1s	95.5 tps
Kimi K2.6 Thinking INT4 moonshotai/kimi-k2.6:thinking:lilac	256k	$0.73/M	$3.68/M	$0.17/M	N/A	3.1s	95.5 tps
MiniMax M3 FP8 minimax/minimax-m3:lilac	512k	$0.29/M	$1.16/M	$0.05/M	N/A	N/A	N/A
MiniMax M3 Thinking FP8 minimax/minimax-m3:thinking:lilac	512k	$0.29/M	$1.16/M	$0.05/M	N/A	N/A	N/A

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 Lilac via the API

For models with provider selection, append :lilac to the model ID, set "provider": "lilac" in the request body, or send an X-Provider: lilac 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": "google/gemma-4-31b-it:lilac",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

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