
Can your favorite model read the fine print?
Signed contracts bury their billing terms in pages of prose. Explore how models compare, then try contract extraction yourself.
python -m fineprint.market_dataRefresh Flexprice Model Intelligence locally from public model rankings. Open source on GitHub.
Model providers
What brought FinePrint to life
Every invoice begins as a sentence in a contract. Read the fee, the cycle or the currency wrong and the bill goes out wrong. The best model still misses a fifth of them.
Contract evaluation methodology
How FinePrint’s evaluation works.
- Contract in
Real PDF, numbered OCR lines.
01 - Model reads it
Structured billing schema.
02 - Private key check
Each field vs the answer key.
03 - Published benchmark
The original accuracy, cost, and speed results.
04
Signed contracts from public filings: order forms, MSAs, renewals. Scanned pages and redlines included. Nothing synthetic.
Every field is hand-labeled. Those labels never ship; they define the FinePrint methodology and keep the extraction evaluation honest.
For scoring details, see our methodology.
Flexprice Model Intelligence
A clearer view of the model landscape.
| / | ||||
|---|---|---|---|---|
| 1Anthropic: Claude Sonnet 5.5 | 96.9 | 92.2 | 82.3 | 91.6 |
| 2Anthropic: Claude Fable 5.1 | 91.9 | 74.1 | 85.9 | 88.3 |
| 3OpenAI: GPT-6 Astra | 90.6 | 81.4 | 89.7 | 87.0 |
| 4DeepSeek: DeepSeek V4.1 Flash | 65.4 | 94.2 | 100.0 | 84.1 |
| 5OpenAI: GPT-5.6 Luna | 60.4 | 96.1 | 98.5 | 82.7 |
| 6OpenAI: GPT-6.1 Sol | 89.0 | 73.7 | 82.0 | 81.9 |
| 7Meta: Muse Spark 1.3 | 81.9 | 74.6 | 84.6 | 81.8 |
| 8OpenAI: GPT-5.6 Sol | 60.4 | 84.3 | 91.1 | 81.1 |
| 9MoonshotAI: Kimi K3 | 73.3 | 82.9 | 90.7 | 81.0 |
| 10OpenAI: GPT-5.6 Terra | 70.4 | 80.5 | 86.2 | 80.6 |
| 11Meta: Muse Spark 1.3 Contributor | 60.5 | 84.1 | 91.5 | 80.0 |
| 12OpenAI: GPT-6 Luna | 62.9 | 89.2 | 92.1 | 79.4 |
| 13Google: Gemini 2.5 Flash Lite | 57.7 | 94.9 | 87.6 | 78.2 |
| 14Qwen: Qwen3.8 Max (0902) | 76.8 | 70.9 | 84.7 | 76.7 |
| 15Google: Gemini 2.5 Flash | 57.7 | 90.5 | 86.1 | 76.3 |
| 16Google: Gemini 3 Flash Preview | 57.7 | 89.0 | 88.1 | 76.3 |
| 17Anthropic: Claude Opus 5.5 | 100.0 | 40.1 | 88.3 | 76.1 |
| 18Google: Gemini 3.1 Flash Lite | 57.7 | 90.0 | 86.2 | 76.1 |
| 19Z.ai: GLM 5.3 | 75.5 | 64.7 | 93.2 | 76.1 |
| 20Xiaomi: MiMo-V2.5 | 53.9 | 87.4 | 95.0 | 75.9 |
Try it
Read a contract with any model.
Try one of the top models live on our samples, or upload your own contract. We run extraction in real time (it can take a minute depending on the model) and show you every field it pulls out.
Read the contract on the left, then Run extraction to see the structured billing schema — every field cited back to the page.
Model metrics are refreshed regularly using a consistent scoring method.
A Note From The Team
Flexprice runs usage-based billing. Before any invoice exists, someone has to turn a signed contract into structured billing terms: the fees, the cadence, the currency, the commitments. We do that step in production.
That experience shapes what we look for in models. We know which fields decide an invoice and how much a missed or misread term matters to the people using it.
FinePrint brings together model rankings with a consistent scoring method, while the scan lets you try an extraction on your own document. If your contracts look different, try the scan on your own.
Flexprice team
Fix your billing today with Flexprice
FinePrint finds which models can read the contract.
Flexprice already turns those terms into accurate invoices.