Two API Settings Triple GPT-5.6's ARC-AGI-3 Score While Cutting Tokens
gabrielchua · x · 2026-07-30
Developers @ilanbigio and @sandersted improved GPT-5.6 Sol’s ARC-AGI-3 score by 3x while using 6x fewer output tokens, simply by adjusting two Responses API settings.
These settings mirror the production setup used in ChatGPT and Codex:
- Retained reasoning: The model remembers its earlier thinking across turns instead of starting from scratch.
- Server-side compaction: Long conversations are compressed while preserving the essential context needed for the model to keep working.
Developers building in Codex are advised to use the openai-docs skill to implement both and maximize the model's value.
Related event: GPT-5.6 Scores Triple on ARC-AGI-3 After Enabling Two API Settings(18 posts)→
More from coding & agent
- Frontier AI Agents Can Do Coding, But Fail at Open-Ended AI Research — Peter Kirgis · 2026-07-30
- Compiling Fuzzy Functions Directly into Neural Weights: The ProgramAsWeights Paradigm — weichiuma · 2026-07-30
- Prediction: AI Models in 2.5 Years to Be 5x Faster, 2x Cheaper, and Approaching Saturation — OfirPress · 2026-07-30
- Deep Dive: Best Technical Routes and Practices for AI-Generated Native PPTs — dotey · 2026-07-30
- Voice Input Reshapes Agent Interaction: Dev Tests $40 Mic for Wispr Flow — edgarpavlovsky · 2026-07-30
- Clarifying Agent Concepts: Differences Between MCP, Skill, and Tool Call — yangyi · 2026-07-30