Data · best for
Best AI model for JSON Extraction (2026)
Pulling structured fields out of unstructured text. Ranked from 343 live models on the OpenRouter catalog, weighted for structured output, low latency, low cost.
| # | Model | Score | In / 1M | Out / 1M | Context | |
|---|---|---|---|---|---|---|
| 1 | Google: Gemma 4 26B A4B (free)google/gemma-4-26b-a4b-it:free | 132 | Free | Free | 262,144 | Try → |
| 2 | Google: Gemma 4 31B (free)google/gemma-4-31b-it:free | 132 | Free | Free | 262,144 | Try → |
| 3 | Google: Gemma 4 26B A4B google/gemma-4-26b-a4b-it | 131 | $0.07 | $0.35 | 262,144 | Try → |
| 4 | Google: Gemma 4 31Bgoogle/gemma-4-31b-it | 131 | $0.13 | $0.38 | 262,144 | Try → |
| 5 | Qwen: Qwen3.5-9Bqwen/qwen3.5-9b | 131 | $0.10 | $0.15 | 262,144 | Try → |
| 6 | ByteDance Seed: Seed-2.0-Minibytedance-seed/seed-2.0-mini | 131 | $0.10 | $0.40 | 262,144 | Try → |
| 7 | Qwen: Qwen3.5-Flashqwen/qwen3.5-flash-02-23 | 131 | $0.07 | $0.26 | 1,000,000 | Try → |
| 8 | ByteDance Seed: Seed 1.6 Flashbytedance-seed/seed-1.6-flash | 131 | $0.07 | $0.30 | 262,144 | Try → |
| 9 | xAI: Grok 4.1 Fastx-ai/grok-4.1-fast | 131 | $0.20 | $0.50 | 2,000,000 | Try → |
| 10 | Google: Gemini 2.5 Flash Lite Preview 09-2025google/gemini-2.5-flash-lite-preview-09-2025 | 131 | $0.10 | $0.40 | 1,048,576 | Try → |
| 11 | xAI: Grok 4 Fastx-ai/grok-4-fast | 131 | $0.20 | $0.50 | 2,000,000 | Try → |
| 12 | OpenAI: GPT-5 Nanoopenai/gpt-5-nano | 131 | $0.05 | $0.40 | 400,000 | Try → |
| 13 | Google: Gemini 2.5 Flash Litegoogle/gemini-2.5-flash-lite | 131 | $0.10 | $0.40 | 1,048,576 | Try → |
| 14 | OpenAI: GPT-4.1 Nanoopenai/gpt-4.1-nano | 131 | $0.10 | $0.40 | 1,047,576 | Try → |
| 15 | Google: Gemini 2.0 Flash Litegoogle/gemini-2.0-flash-lite-001 | 131 | $0.07 | $0.30 | 1,048,576 | Try → |
How we ranked these
For JSON Extraction, we weight models on structured output, low latency, low cost. Higher means better. Scores combine OpenRouter's model metadata (context length, modality support, tool calling, structured output, reasoning capability) with public pricing. See full methodology →
Related tasks
Data
Best for Data Analysis
Exploring datasets, drawing conclusions, computing summary stats.
Data
Best for CSV / Spreadsheet Cleanup
Normalizing messy tabular data with consistent fields.
Data
Best for ETL Scripting
Writing extract-transform-load pipelines.
Data
Best for Bulk Data Labeling
Cheaply tagging thousands of items with consistent labels.
Data
Best for OCR / Document Parsing
Reading text out of images, PDFs, and scanned documents.
Data
Best for Table Extraction from PDFs
Pulling structured tables out of complex documents.