OPEN LINK ↗
// 1h agoBENCHMARK RESULT
GPT-6.1 Terra takes MacroscopeBench’s No. 2 spot
GPT-6.1 Terra ranks second on Macroscope’s real-world code-review benchmark, trailing Claude Opus 5.5 max by just 1.6 F1 points at roughly one-twentieth the cost per review. At medium effort, it reportedly surpasses GPT-6.1 Sol max for about one-third the price.
// ANALYSIS
Terra’s standout advantage is not absolute quality but an unusually strong quality-to-cost curve for automated review pipelines.
- –MacroscopeBench evaluates bug detection on real code changes, measuring recall against precision rather than code-generation ability.
- –The near-Opus result suggests Terra could handle first-pass review, CI checks, and routine pull requests at much higher volume.
- –Beating GPT-6.1 Sol max at medium effort makes Terra especially compelling where latency and review cost matter.
- –Macroscope’s benchmark is vendor-published, so independent replication and larger task coverage remain important before treating the ranking as definitive.
- –OpenAI positions Terra as its efficient model tier, while reserving Sol and Astra for the most demanding coding work.
// TAGS
gpt-6.1-terrallmevaluationbenchmarkcode-reviewai-codingcoding-agent
DISCOVERED
1h ago
2026-10-31
PUBLISHED
1h ago
2026-10-31
RELEVANCE
9/ 10
AUTHOR
Macroscope