GPT-6 Luna trails its predecessor 5.6 on several benchmarks; the real win is price
DataLearnerAI · reddit · 2026-09-23
While updating benchmark data, DataLearnerAI noticed that GPT-5.6 Luna actually outscores GPT-6 Luna on quite a few reported benchmarks. Cross-checking Artificial Analysis, the overall Intelligence Index at max reasoning is basically a tie: 37 vs 37.
Individual results are mixed: GPT-6 Luna improves on some agent/coding tasks while 5.6 still wins several others. The clearest gain is efficiency — GPT-6 Luna is substantially cheaper.
The author speculates GPT-6 Luna may be a deliberate cost/speed optimization rather than a straight capability upgrade, or that harness/reasoning-setting differences make the comparisons unfair, and asks what others see in real usage.
More from Models
- Opus 5.5 stop-motion demos look alike, raising questions on creative heterogeneity — dylfreed · 2026-09-23
- New Paper: Transformers Hold Faithful Internal World Maps but Fail to Act on Them — xuanalogue · 2026-09-23
- GPT-6 Luna Appears; PSAISuite Works With Zero Changes — dfinke · 2026-09-23
- OpenAI Reportedly Expands GPT-6 Lineup With Sol and Luna Models — dfinke · 2026-09-23
- Rumor: Opus 5.5 Said to Be Cheaper and Faster Than Astra — BLUECOW009 · 2026-09-23
- The industry dropped the ball on strong, tool-calling, non-reasoning small LLMs — vboykis · 2026-09-23