Raw text input
AI Models & Assistants / Knowledge library
GPT-4
An older high-intelligence GPT model
Why learn GPT-4?
Offers the documented OpenAI model capabilities through hosted.
The problem it solves
Provides model inference for supported text inputs.
Where it came from
GPT-4: An older high-intelligence GPT model Published aliases: gpt-4-0613, gpt-4-0314. Input: text. Output: text. Checked 2026-10-13. Model capabilities are distinct from tools available in a particular agent.
Build your understanding
Work through these exercises in order. Try the task before opening a reference.
- Explain what GPT-4 does.
- Identify a use case for GPT-4.
- Describe a limitation of GPT-4.
- Name a compatible tool for GPT-4.
- Compare GPT-4 against an alternative.
Documentation and sources
- Primary source / model card (developers.openai.com)
Catalog research notes
- See the AI Workbench for source dates, pricing evidence and configuration details.
Connect the ideas
Built on
Real World Example
Automated Legal Document Summarization for Boutique Law Firms
Boutique law firms handling high volumes of discovery documents.
The business problem
Attorneys struggle with the cognitive load of manually reviewing hundreds of pages of case discovery to extract key arguments, leading to potential oversight of critical details.
How this topic solves it
Utilizing the GPT-4 model via the Chat Completions API, firms can automate the extraction of core legal arguments and entity relationships from text-based discovery documents, providing a structured summary for human review.
- Ingest raw text discovery documents into a secure processing pipeline.
- Format the text into structured prompts for the gpt-4-0613 snapshot to ensure consistent output behavior.
- Send requests to the v1/chat/completions endpoint with specific instructions to identify parties and claims.
- Parse the returned text output to populate a case management database for attorney review.
This diagram illustrates the flow of data from raw discovery files to the final attorney review interface.
GPT-4 Chat Completions
Extracted arguments
Final decision making
- Discovery Docs → Ingestion → API Request
- API Request → Generation → Structured Summary
- Structured Summary → Validation → Attorney Review
Expected business outcome
Improved efficiency in document triage, allowing attorneys to focus on high-level legal strategy rather than initial data extraction.
Limitations and checks
The 8,192 token context window limit restricts the size of individual documents that can be processed in a single request; check for truncation by monitoring the finish_reason in the API response.