Kimi K3 Leads Multi-Sampling Benchmarks
zephyr_z9 · x · 2026-07-18
The image compares the performance of Kimi K3 and Fable 5 across different sampling attempts:
- pass@1: Kimi K3 scores 68.5, slightly below Fable 5's 69.9.
- pass@2: Kimi K3 scores 82.0, beating Fable 5's 80.2.
- pass@4: Kimi K3 scores 89.4, ahead of Fable 5's 88.5.
The post highlights that Kimi K3 performs significantly better under multi-sampling settings, earning it the title of "open/closed SOTA".
More from Models
- Developer Building a Unified Leaderboard of All Model Benchmark Scores — airesearch12 · 2026-09-11
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11
- User Hails Unconfirmed 'DeepSeek 4.1 Flash' as an Inflection Point in LLMs — himanshustwts · 2026-09-11
- Terminal Bench v4: GLM-5.3 Leads at 41.9%, Kimi-K3 Underwhelms at 12.6% — Ok_Warning2146 · 2026-09-11