Kimi-K3 Review: Top Open-Source but Costly
scaling01 · x · 2026-07-18
In the LisanBench evaluation, Kimi-K3 emerged as the strongest open-weight model currently available, ranking 5th in standard metrics and 4th in difficulty-weighted metrics.
- Performance: It outperformed Gemini 3.1 Pro (high), GPT-5.5, and Opus 4.8, though it still falls short of Opus 4.7 xhigh at comparable token usage.
- Cost: Despite its excellent performance, it consumes 2-3 times more tokens than the compared models, making actual usage costs significantly higher.
Related event: Kimi-K3 Tops LisanBench as Strongest Open-Weight Model(2 posts)→
More from Models
- Claim says Kimi was distilled from Fable, sparking a model-attribution jab — cephaloform · 2026-07-22
- Gemini 3.6 Flash is now available in Antigravity and chat — MartianOnJupiter · 2026-07-22
- Gary Marcus says LLM math skills are like knowing only a car’s engine size — GaryMarcus · 2026-07-22
- OpenAI’s Codex + GPT-5.6 Sol hits 99% recall in Project APE verification tests — soumitrashukla9 · 2026-07-22
- OpenAI-linked paper says capability RL can make models more reward-seeking — MariusHobbhahn · 2026-07-22
- Macaron V1 adds LoRA RL on GLM 5.2 and claims SOTA benchmark gains — Xianbao_QIAN · 2026-07-22