# Inference.net: Schematron V2 Small

Provider: inference-net  
Category: llm  
Model ID: `inference-net/schematron-v2-small`

Schematron V2 Small is a 3B-parameter HTML-to-JSON extraction model from Inference.net. It prioritizes extraction quality for complex schemas and long pages. Extraction instructions must be supplied through a JSON schema...

## Specs

- Context length: 128000 tokens
- Max output: 4096 tokens
- Modalities: text
- Released: 2026-09-12

## Pricing

- Input: $0.05 per million tokens
- Output: $0.23 per million tokens

## Providers

- **inference-net** — ctx 128000, input $0.05/M, output $0.23/M

## Verdict

Inference.net's Schematron V2 Small offers a 128K context window at budget pricing — $0.05 input and $0.23 output per million tokens. Without public benchmarks, it's hard to gauge performance against peers like GPT-4o-mini or Gemini Flash. The pricing suggests this is positioned as a cost-efficient workhorse for high-volume tasks where you need long context but can tolerate less polish than frontier models. Best for teams running cost-sensitive workloads who can validate quality in their own domain before committing.

---
Last verified: 2026-10-19T02:30:34.239Z  
Canonical URL: https://switchy.build/directory/models/schematron-v2-small