Benchmarkers slam Kimi K3.1 Pro pricing: 53.7% score for $31.88, cache-read games suspected
natolambert · x · 2026-10-24
Benchmarker aicodeking tested Kimi K3.1 Pro on his KingBench 4 and found the 'cheap' pitch unconvincing: 53.7% (6th place) at $31.88 per run, only $16 cheaper than Opus — Haiku or Sol deliver better results $25 cheaper. He claims the model constantly re-reads cache, inflating costs while cache reads are priced the same as Opus, speculating this is a way to recover training costs. natolabert amplified it, noting few positive reactions on the timeline, and added that at the top end, real-world usability matters more than ±2 points on leaderboards — K3.1 may be on the 'miss' side, even as OpenAI dropped out of AA Analysis' top 5 after its release.
More from Models
- Leaked Claude UI screenshot shows unannounced Opus 5.5, Opus 5 and Fable 5.1 models — adonis_singh · 2026-10-24
- Developer on Claude Opus 5.5: "nothing else even comes close right now" — prasenx · 2026-10-24
- NYU/UT Austin paper: LLMs write convoluted text due to length-penalty optimization — menhguin · 2026-10-24
- 50 pre-ChatGPT novels tested against Pangram AI detector: zero false positives — TuhinChakr · 2026-10-24
- Seven independent tests find AI detector Pangram false-positive rate near zero — TuhinChakr · 2026-10-24
- Dev compiles incident-triage SOP into a fine-tuned 3B model: 93% accuracy at 8.4x lower cost — arpit_bhayani · 2026-10-24