發布日期2026/07
推論速度116 tokens/s
首字延遲時間74.83s
輸入價格$2/1M
輸出價格$12/1M
排名表現
評測分析
GPT-5.6 Terra is a top choice for developers and researchers when coding and reasoning quality take priority over price or speed. Not suited for automation tasks or simple chatbots, it performs best on tasks requiring high precision in programming and analysis.
強項
- Extremely strong coding, ranked #2 globally—handles high code complexity, precise debugging, fast algorithm optimization
- Superior logical reasoning, ranked #3—analyzes complex mathematical problems with clear step-by-step deduction
- Broad, deep knowledge base suitable for technical conversation—nuanced, non-simplistic answers
弱項
- Weaker autonomous capabilities (AI Agent)—ranked #15, better suited for cognitive work than automated execution
- High operational cost; inference speed may not be ideal for low-latency applications
- Not optimized for small, repetitive tasks—best reserved for brain-heavy technical challenges
適用情境
Assisting software engineers with writing, reviewing, and refactoring complex codeSolving scientific, mathematical, and logic problems requiring detailed step-by-step reasoningServing as a sparring partner for AI research and deep scientific analysis—where reasoning quality is the top priority
指南與影片
Access GPT-5.6 Terra via ChatGPT.com (web interface) or the OpenAI API with model ID 'gpt-5.6-terra'. This model is designed for tasks requiring a balance between performance and cost, such as customer support, internal tools, and document analysis. Priced at $2.50 input / $15 output per 1M tokens—50% cheaper than Sol while more powerful than Luna. Supports a 1.05M-token context window, reasoning levels (low/medium/high/xhigh/max), and prompt caching.