OpenAI GPT-6 Astra is OpenAI's most capable model, built for the hardest end-to-end work — complex reasoning, coding, computer use, research, and document creation — with a 1,050,000-token context window and up to 128,000 tokens of output. Stronger reasoning, coding, multimodal understanding, and long-form writing than GPT-5.6 Sol. Supports text and image input, web search, and reasoning effort (five levels from low to max; this model has no 'none' setting) / summary display options. Best for large codebases, cross-document research, professional reports, and multi-step reasoning; it costs more than GPT-5.6 Sol, so for everyday tasks GPT-5.6 Terra or Luna is cheaper. If you need to search the web for information, please turn on the "Web Search" switch. Prompt example 1: (Upload three annual financial reports, PDF) Compare revenue, gross margin and cash flow across three years, flag anomalies and causes, and write a report with tables. Prompt example 2: Design a multi-warehouse inventory system with database schema, API specs, a restocking algorithm and a test plan, then write full code for the core restocking module. Prompt example 3: Plan a one-year carbon inventory rollout for a mid-sized manufacturer: steps, department roles, data collection forms, risk responses and a slide outline for management. Prompt example 4: (Upload five competitors' product catalogs, PDF) Build a feature and spec comparison matrix, find unmet market needs, and draft a launch strategy for Northwind Ltd.'s new product. Prompt example 5: Our ten-year-old Java monolith for orders must become microservices. Plan service boundaries, data migration, running old and new side by side, rollback, and staffing per phase. Prompt example 6: Derive the orbital decay equation for a low-orbit satellite under atmospheric drag, estimate when a small satellite at 400 km reenters, state your assumptions, and verify with a simulation. Prompt example 7: (Turn on web search) Review solid-state battery mass-production progress over the past three years, compare the main technical approaches and bottlenecks, and write a research brief with sources. Prompt example 8: (Upload a results chart, png, and a paper draft, docx) Check whether the chart data supports the paper's conclusions, point out statistical issues, and suggest follow-up experiments. Prompt example 9: (Upload a draft purchase agreement, docx) Review each clause for balance between the parties, flag terms unfavorable to us, and propose revised wording and negotiation priorities. Prompt example 10: Write a complete employee handbook for a 50-person software company covering onboarding, attendance, leave, performance reviews, information security and offboarding, with an FAQ per chapter. Maximum context window (input and output combined): about 1,050,000 tokens. Maximum total input per request (including history): 922,000 tokens. Maximum output per reply: about 128,000 tokens (model reasoning, if any, counts toward this limit). Maximum total upload size per request: 512MB (all files and message content combined). Accepted document formats: pdf, docx, xlsx, pptx, txt, csv (file content is provided to the model as extracted text). Accepted image formats: gif, jpeg, jpg, png, webp. Maximum number of images: 1500. Maximum size per image: 50MB. Maximum total image size per message: 384MB (text and history share the same request size limit; uploads close to the limit may still be rejected). Image dimensions: 200 to 8,000 pixels per side. Maximum image resolution: 36,000,000 pixels in total. Output content formats: Text/Image Pricing: Very high Fast Mode: uses 2x credits Continuous conversation: Supported AI can make mistakes. Please verify important information.