Coverage (30d)
0vs17
This Week
0vs4
Evidence
1 articlesRelationships
1Timeline
GPT-5.32026-03-26
Achieved 100% resident identification accuracy in a safety evaluation for a care home smart speaker system.
GPT-5.32026-03-07
Released as OpenAI's most capable frontier model with unified coding, reasoning, and computer operation capabilities
GPT-5.32026-03-06
Demonstrated surpassing human baselines on OSWorld benchmark with 75% score
GPT-5.32026-03-05
OpenAI releases GPT-5.4 with native computer use, tool search, and 1M token context window
GPT-5.32026-03-01
Expected to follow shortly after DeepSeek v4 release
GPT-5.32026-02-28
Identified as experiencing up to 33% accuracy degradation in extended conversations according to new research
Evolver2026-02-28
Achieved 95% accuracy on ARC-AGI-2 benchmark
Ecosystem
Evolver
competes withGPT-5.31 src
usesARC-AGI-21 src
GPT-5.3
developed byOpenAI
developedOpenAI5 src
usesPlaywright1 src
usesRust1 src
competes withClaude Sonnet 4.61 src
usesGPT-5.3-Codex1 src
Benchmarks
gpqa
Evolver—
GPT-5.392
swe bench pro
Evolver—
GPT-5.356.8