1. Strongest Agentic Coding Model

• State-of-the-art on Terminal-Bench 2.0 (82.7%), Expert-SWE (73.1%), and SWE-Bench Pro (58.6%), outperforming GPT-5.4 on complex coding tasks.
• Holds context across large systems, reasons through ambiguous failures, and carries changes through surrounding codebases with fewer tokens.

2. Higher Intelligence at GPT-5.4 Latency

• Co-designed, trained, and served on NVIDIA GB200/GB300 NVL72 systems to match GPT-5.4 per-token latency while performing at a significantly higher level.
• Uses fewer tokens to complete the same tasks, making it more efficient as well as more capable.

3. Powerful for Knowledge Work & Computer Use

• Scores 84.9% on GDPval (44 occupations) and 78.7% on OSWorld-Verified for autonomous computer operation.
• Excels at generating documents, spreadsheets, and reports; naturally moves across finding information, using tools, and checking output.

4. Scientific Research Co-Scientist

• Leading performance on GeneBench, BixBench, and FrontierMath; helped discover a new proof about Ramsey numbers verified in Lean.
• Strong enough to meaningfully accelerate progress at the frontiers of biomedical and mathematical research.

1. Configurable Reasoning for Agentic Tasks

• Built to excel in autonomous or semi-autonomous coding workflows, with adjustable reasoning effort for planning, refactoring and debugging.

2. Fast Multi-Modal Input with Large Output

• Accepts both text and image inputs while producing text outputs.
• Offers up to 128 k output tokens, allowing long responses and code generation across multiple files.

3. Large Context & Knowledge Cut-Off

• 400 k token context window supports processing large codebases or documents.
• Knowledge cut-off of Sep 30 2024 ensures familiarity with recent tools and frameworks.

4. Reasoning Token Support

• Provides explicit support for reasoning tokens, enabling developers to fine-tune the balance between reasoning depth and speed.