Kimi K3 beats Inkling on five benchmarks, but the size gap is huge
echen · x · 2026-07-24
A comparison chart shows Kimi K3 versus Thinking Machines' Inkling across several benchmarks.
- The author notes the comparison is not apples-to-apples because K3 has 2.8T parameters while Inkling has 975B.
- The chart highlights K3 ahead on all shown benchmarks, including:
- Antidote: 1077 vs 985 Elo
- HANDBOOK.md: 11.9% vs 1.9%
- Chartography: 26.6% vs 3.4%
- Riemann-bench: 37.6% vs 15.2%
- ComplexConstraints: 37.9% vs 0.3%
The post is basically a benchmark snapshot with a size caveat, not a full technical analysis.
Related event: Kimi K3 Outperforms Inkling in Benchmarks Despite Size Disparity(2 posts)→
More from Models
- Anthropic's Terra returns with better writing and 2X usage vs 6.1 Sol — weswinder · 2026-10-27
- User catches model failing to parse 400-paper file, citing stale news sources instead — jungofthewon · 2026-10-27
- Odd eval result: Gemini 4.1 Flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Poster surprised that v4.1 flash gets more accurate when asked to control its CoT — dfrsrchtwts · 2026-10-27
- Dev claims Anthropic is baking 'sabotage' into model behavior for 'wrong' engineering — cephaloform · 2026-10-27
- Gemini fails at listing unread emails, tells user to search Gmail manually — gethackteam · 2026-10-27