inference-net · Efficient
Schematron V2 Turbo
Inference.net's 3B-parameter HTML-to-JSON extraction model, optimized for throughput and low cost on high-volume workloads. It turns HTML into typed, structured data for web scraping and product catalog ingestion, with a 128K-token context window. Supply HTML in the user message and extraction instructions in a JSON schema via response_format; it does not follow ordinary chat or system prompts.
- Structured output
- Input
- Text
- Output
- Text
- Discovery tier
- Efficient
Pricing
USD per 1M tokens- Input
- $0.03
- Output
- $0.15
Reference prices as published by the model provider. API example
Specifications
- Context window
- 128K
- Max output
- 8.2K
Side by side
Schematron V2 Turbo compared
Specifications, prices, speed and benchmark scores next to each other. Benchmarks below show every compared model too.
| Schematron V2 Turboinference-netThis model | Solar Pro 4upstageRemove | |
|---|---|---|
| Provider | inference-net | upstage |
| Discovery tier | Efficient | Efficient |
| Added | Sep 12, 2026 | Aug 11, 2026 |
| Limits | ||
| Context window | 128K | 524.29KBest |
| Max output | 8.19K | 131.07KBest |
| Average output | — | 546 tokens |
| Pricing, USD per 1M tokens | ||
| Input | $0.03 | $0.03 |
| Output | $0.15 | $0.12Best |
| Cache read | — | $0.006 |
| Performance | ||
| Throughput | — | 123.6 tok/s |
| Time to first token | — | 4.5 s |
| Request success (24h) | — | 99.9% |
| Benchmark indexes | ||
| Intelligence Index | — | 26.6 |
| Capabilities | ||
| Tools | No | Yes |
| Structured output | Yes | Yes |
| Reasoning effort | — | — |
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
- Schematron V2 Turbo
- Solar Pro 4
Intelligence Index
- Schematron V2 Turbo—
- Solar Pro 426.6
Agentic work
AutomationBench-AA
Workflow automation, penalized for guardrail violations
- Schematron V2 Turbo: —
- Solar Pro 4: 13.6%
AutomationBench-AA tasks completed
Workflows finished end to end without a guardrail violation
- Schematron V2 Turbo: —
- Solar Pro 4: 2.4%
AA-Briefcasev1.1
Agentic knowledge work, rated head to head
- Schematron V2 Turbo: —
- Solar Pro 4: 1,093 Elo
GDPval-AAv2.1
Economically valuable occupational tasks, rated head to head
- Schematron V2 Turbo: —
- Solar Pro 4: 1,106 Elo
Document reasoning
GDP.pdf
Professional reasoning over PDF documents, all-pass rate
- Schematron V2 Turbo: —
- Solar Pro 4: 7.0%
AA-LCRv1.1
Reasoning across long documents
- Schematron V2 Turbo: —
- Solar Pro 4: 70.7%
Reasoning and knowledge
Humanity’s Last Exam
Expert-level questions across academic fields
- Schematron V2 Turbo: —
- Solar Pro 4: 29.8%
GPQA Diamond
Graduate-level science questions
- Schematron V2 Turbo: —
- Solar Pro 4: 89.1%
Coding
Terminal-Benchv4.0
Practical software engineering in a terminal
- Schematron V2 Turbo: —
- Solar Pro 4: 0.0%
SciCode
Scientific computing in Python
- Schematron V2 Turbo: —
- Solar Pro 4: 48.0%
Legacy evaluations
AA-LCR (unversioned)
Long-context reasoning, earlier grading
- Schematron V2 Turbo: —
- Solar Pro 4: 70.7%
Scores from Artificial Analysis, updated Oct 26, 2026. Artificial Analysis
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "inference-net/schematron-v2-turbo",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="inference-net/schematron-v2-turbo",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)inference-net
Related models
Other inference-net models, starting with the same family.
- EfficientCompare