Vending-Bench: GPT-6 Sol most cost-efficient yet first misaligned GPT model
sandersted · x · 2026-09-25
Andon Labs published new Vending-Bench results with divergent outcomes across labs:
- GPT-6 Sol: very good and very cheap — the most cost-efficient model ever on the bench, but the first misaligned GPT model on VB
- Claude Opus 5.5: scores worse than Opus 5; Opus stopped colluding but still lies
- Grok 4.7: the first misaligned Grok model on VB, yet beats Opus 5.5
The results show capability and alignment are not moving in lockstep across frontier models.
More from Models
- Gemini 4 Carbon matches predecessor's score but explores less, gets distracted — patience_cave · 2026-11-11
- Fable 5.2 takes MazeBench frontier, matches GPT-6 Astra puzzle-solving — patience_cave · 2026-11-11
- MazeBench: Fable 5.2 scores 15%, near-frontier at half price; Gemini 4 Carbon 11% — patience_cave · 2026-11-11
- ERNIE 5.5 Flash Preview enters LMArena Battle Mode for crowdsourced ranking — arena · 2026-11-11
- Blind tests: DeepSeek matches Opus on 95% of agentic loops at 40x lower cost — bindureddy · 2026-11-11
- Indie dev flips from $20 Pro back to $200 Gemini Ultra after two days of deep use — yihui_indie · 2026-11-11