plugsky-max — heavy reasoning, zero drama

Summary

Model	Identity	Context	Latency*	Best for
plugsky-max	Qwen 3.5-class (MoE)	256K tokens	~0.71s median	Deep reasoning, multimodal, long documents, agents

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

Specs

Property	Value
Model family	Qwen 3.5-class (MoE, 397B)
Context window	256K tokens
Measured latency	~0.71s median on the live API
Capabilities	Reasoning, multimodal, tool calling
Available on	Builder plan and above

When to use plugsky-max

• Deep reasoning — the hardest prompts your product sees.
• Multimodal — images plus text in one call.
• Long-document analysis — 256K context for contracts, papers, logs.

FAQ

Does plugsky-max accept images?

Yes — multimodal input is supported. Dedicated vision models are also available (e.g. plugsky-vision-fast).

Which plan do I need?

Builder and above include plugsky-max access.