Kimi K3 and Fable 5 now look much closer than the old open-vs-closed gap
FinanceYF5 · x · 2026-07-21
This post frames a deeper look at Kimi K3 versus Fable 5 on software engineering / DeepSWE tasks and argues that frontier open models are no longer six months behind. The chart shown in the image is a side-by-side product card: Kimi K3 is open source, while Fable 5 is closed.
The thread’s surrounding results suggest Kimi K3 is now competitive with top closed models on best-of-k coding-style tasks, with only small gaps in pass rates and failure patterns that look increasingly similar across vendors.
More from Models
- Moonshot pauses Kimi K3 signups five days after launch as GPU demand surges — eyishazyer · 2026-07-21
- AI Diplomacy demo makes agents negotiate, ally, and betray each other — jamdac · 2026-07-21
- Newer models need a different prompting style, and old tricks can make outputs worse — emollick · 2026-07-21
- GLM-5.5 is said to arrive in 4 weeks with open weights — tanay_mehta · 2026-07-21
- Fable 5 is credited with a 3-variable counterexample to the Jacobian conjecture — Various-Affect4841 · 2026-07-21
- Ben’s Bites roundup highlights Kimi K3, Fable 5, Cursor costs and self-driving companies — Ben's Bites · 2026-07-21