plugsky-lite — speed without the tradeoff

Summary

Model	Identity	Context	Latency*	Best for
plugsky-lite	Llama 3.2 3B	128K tokens	~0.24s median	High-volume chat, classification, extraction, low-latency agents

*Median measured latency on the live API at https://api.plugsky.com/v1.

Specs

Property	Value
Model family	Llama 3.2 (3B)
Context window	128K tokens
Measured latency	~0.24s median on the live API
Capabilities	Tool/function calling, streaming, long context
Available on	Free plan and above

When to use plugsky-lite

• High-volume traffic — support bots, classification, extraction.
• Cost-sensitive workloads — smallest model that passes your eval.
• Edge-latency needs — interactive UX that cannot wait a second.

Use it

curl https://api.plugsky.com/v1/chat/completions \
  -H "Authorization: Bearer sk-live-..." \
  -H "Content-Type: application/json" \
  -d '{"model":"plugsky-lite","messages":[{"role":"user","content":"hi"}],"stream":true}'

FAQ

How fast is it really?

Median ~0.24s in live measurements on the public API.

Is it accurate enough for production?

For structured, high-volume tasks — yes. For deep reasoning, pair it with plugsky-micro for hard prompts.