Inference.net models

Inference.net builds the Schematron models, small models trained for one job: turning HTML into structured JSON against a caller-supplied schema. Zeplik carries both V2 servings — Turbo, tuned for throughput on high-volume extraction, and Small, tuned for extraction quality on complex schemas and long pages.

Ask a Inference.net model now

All 2 Inference.net models

Pricing at a glance

Frequently asked questions

Which Inference.net models can I use on Zeplik?

Zeplik carries 2 Inference.net models: Schematron V2 Turbo and Schematron V2 Small. They all run in the same chat, so you do not need a separate Inference.net account or API key.

How much do Inference.net models cost on Zeplik?

Output tokens run from 1.8 to 2.8 credits per million depending on which Inference.net model you pick (raw provider rates $0.15 to $0.23 per million). 1 credit = $0.10, computed from the raw provider rate with a 1.20x margin.

Which Inference.net model has the largest context window?

Schematron V2 Turbo, at 128K tokens (128,000). The context window covers the conversation plus any documents you attach.

Do Inference.net models support images and tool calling?

None of the 2 accept image input, and none support tool calling.

Do I need a paid plan to use Inference.net models on Zeplik?

Both need a paid plan (Plus or above). Usage is then billed in credits from your monthly allowance.

What is the newest Inference.net model on Zeplik?

Schematron V2 Turbo, released September 2026. Schematron V2 Turbo is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes throughput for high-volume extraction workloads.

Other providers

• OpenAI (100)
• Qwen (53)
• Google (42)
• Anthropic (29)
• Mistral (26)
• Z.ai (18)
• DeepSeek (15)
• NVIDIA (11)
• All models