GPT-5.3-Codex | Hacker News

kept by eddie feb 06

OpenAI releases GPT-5.3-Codex, a faster agentic coding model that achieves state-of-the-art results on SWE-Bench Pro and Terminal-Bench 2.0.

Demystifying evals for AI agents
Anthropic's guide to building evaluations for AI agents emphasizes structured tasks, multiple grader types, and iterative refinement to enable confident shipping at scale.
also on Terminal-Bench, #benchmarks, Anthropic, #developer-tools
Introducing Composer 2
Cursor released Composer 2, a frontier-level coding model achieving strong benchmarks (CursorBench 61.3, Terminal-Bench 61.7) at competitive pricing.
also on Terminal-Bench, #benchmarks, #ai-coding, #developer-tools
Opus 4.6, Codex 5.3, and the post-benchmark era
The latest coding models from OpenAI and Anthropic show marginal benchmark gains but real-world usability differences, with Claude ahead in product experience while Codex edges in coding capability.
also on gpt-5.3-codex, #benchmarks, OpenAI, Anthropic, #ai-coding, #developer-tools
Introducing GPT-5.3-Codex
OpenAI launches GPT-5.3-Codex, a coding agent that improves on prior benchmarks, runs 25% faster, and was used to develop itself.
also on gpt-5.3-codex, Nvidia, OpenAI, Anthropic, #ai-coding, #developer-tools
Intelligence Yield — METR Time Horizons v1.1
Opus 4.6 delivers 14 times more useful work per compute-minute than Codex 5.3, a metric called Intelligence Yield derived from METR Time Horizons data.
also on Opus 4.6, #benchmarks, #ai-research