GPT-6.2 Astra Law
At a glance
- Price per 1M tokens (input)
- $8.00
- Price per 1M tokens (output)
- $40.00
- Context
- 922,000 tokens
- Free access
- yes (see below)
Refreshed daily; data verified November 20, 2026. Published November 19, 2026.
GPT-6.2 Astra Law by OpenAI is a language model built for legal professionals, compliance teams, and organizations that process structured and unstructured documents. It supports image understanding for scanned contracts and filings, tool calling to connect with external databases and applications, and step-by-step reasoning for multi-stage legal analysis. OpenAI's approach centers on aligning model behavior with safety guidelines while enabling developers to extend capabilities through programmable interfaces.
Specifications & pricing
| Input (per 1M tokens) | $8.00 |
|---|---|
| Output (per 1M tokens) | $40.00 |
| Cache read (per 1M tokens) | $0.40 |
| Cache write (per 1M tokens) | $10.00 |
| Context window | 922,000 tokens |
| Max output | 128,000 tokens |
| Capabilities | images, tool calling, reasoning |
LiteLLM community dataset (MIT), verified November 20, 2026. Official OpenAI pricing.
What GPT-6.2 Astra Law would cost on your workload — run it through the cost calculator →
Where to try GPT-6.2 Astra Law for free
- OpenAI offers a free chat — ChatGPT (free plan). A vendor's free chat may run a different model from the same family — the exact model is not guaranteed.
Frequently asked questions
What tasks is this model suited for?+
It is suited for contract review, legal research, document summarization, and compliance checking. Image understanding lets it read scanned pages, while tool calling lets it query case management systems or other external tools during a task.
How do I get started?+
Access the model through OpenAI's platform, create an API key, and send prompts using the chat completions interface. You can define functions for tool calling so the model can retrieve information from your own systems.
How does it differ from sibling models in the line?+
This variant is tuned for legal language and document-heavy workflows. Sibling models may prioritize general conversation, coding, or creative writing rather than structured legal analysis.
What are its limitations?+
It does not provide licensed legal advice and can miss jurisdiction-specific nuances. Image understanding depends on document quality, and tool calling requires you to build and maintain the external functions it invokes.
Compare with others

Org chart: how to move your company onto AI
A practical map: which company roles and processes AI agents can take over, where to start, and in what order to roll it out.