Schematron V2 Turbo
Schematron V2 Turbo is a large language model made by Inference.net, added to the catalog on 12 Sep 2026. It has a context window of 128,000 tokens and costs $0.03 per 1M input tokens and $0.15 per 1M output tokens. It accepts text and produces text. Typical uses: schema-driven HTML data extraction, high-volume HTML extraction workflows and returning extracted information in structured text.
Price checked ยท source: OpenRouter
What Schematron V2 Turbo is
Inference.net: Schematron V2 Turbo is a text model for extracting information from HTML and returning it as JSON. The provider describes it as a 3-billion-parameter model that prioritizes throughput for high-volume extraction workloads. It accepts text and produces text, with HTML-to-JSON extraction as its stated task. This makes it suited to workflows where information needs to be pulled from HTML and represented in a defined structure.
Extraction instructions must be supplied through a JSON schema in the response_format parameter. The model supports response_format and structured outputs, making schema-driven formatting part of its documented interface. Its use is focused on extracting information according to a supplied schema. The available facts do not specify particular HTML elements, fields, or schema conventions, so those depend on the instructions provided with a request.
The model has a context window of 128,000 tokens and a maximum output of 8,192 tokens. This context limit allows substantial text input to be considered alongside schema instructions, while the output limit sets a ceiling on the size of a response. Its stated focus on high-volume workloads makes it relevant to repeated HTML extraction tasks. The listed input and output modalities are text; the catalog does not list image or other non-text input. The model is therefore most directly suited to text-based extraction requests that specify their desired output structure in advance.
Key features
- Extracts information from HTML into JSON
- Uses a JSON schema in response_format for extraction instructions
- Supports structured outputs and response_format
- Has a 128,000-token context window
- Allows up to 8,192 output tokens
Typical uses
- Schema-driven HTML data extraction
- High-volume HTML extraction workflows
- Returning extracted information in structured text
Schematron V2 Turbo pricing
US dollars, list price through OpenRouter.
| Item | Standard |
|---|---|
| Input, per 1M tokens | $0.03 |
| Output, per 1M tokens | $0.15 |
| Cached input read, per 1M tokens | $0.03 |
No price change since it was added on 12 Sep 2026.
| What a task costs | Cost |
|---|---|
| Answer a short question500 in / 300 out | $0.0001 |
| Summarise a 50-page PDF25,000 in / 800 out | $0.0009 |
| Review a 2,000-line source file30,000 in / 2,000 out | $0.0012 |
| Read a 300-page contract set150,000 in / 1,500 out | $0.0047 |
Schematron V2 Turbo specifications
| Model id | inference-net/schematron-v2-turbo |
|---|---|
| Version | schematron-v2-turbo-20260902 |
| Context window | 128,000 tokensAbout 96,000 English words |
| Max output | 8,192 tokens |
| Knowledge cutoff | Not published |
| Added to catalog | 12 Sep 2026 |
| Input types | text |
|---|---|
| Output types | text |
| Tool calling | No |
| Structured outputs | Yes |
| Reasoning effort control | No |
| Content moderation | Not applied by provider |
Sample outputs
The same prompts are given to every model, and the unedited output is shown with the date it was produced. Upend does not publish leaderboard scores it did not run.
No output has been recorded for Schematron V2 Turbo yet.
Compare Schematron V2 Turbo
Side by side with other models: prices, context window, what each accepts and supports, and the same four tasks priced on both.