Kimi K3 costs $4.65 per rollout but takes 66 minutes versus Sol’s 17
zainhas · x · 2026-07-23
A second chart says Kimi is cheaper per solve, while Sol is much faster per rollout.
Reported numbers:
- Cost per rollout: Kimi $4.65 vs Sol $8.37
- Solved rollouts per $100: Kimi 14.7 vs Sol 8.6
- Median rollout time: Kimi 66 minutes vs Sol 17 minutes
The author notes that Kimi’s cost advantage is clear, but the end-to-end latency should improve once more inference providers support the model. The slower runtime is attributed to model behavior, especially extra steps.
Related event: Kimi K3 Max vs GPT-5.6 Sol Max: A Routing Problem in Coding(10 posts)→
More from Infra
- Resin lets an agent build standalone LLM inference engines; vLLM and SGLang optimize DeepSeek kernels differently — AccBalanced · 2026-10-27
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- Anthropic S-1: CSP marketplace revenue jumped from 11% to 48% in three years — AccBalanced · 2026-10-27
- Alchemy adds effect/Workflow support across Durable Objects, Workflows and Lambda Durable Functions — samgoodwin89 · 2026-10-27
- Alchemy's Lambda Web Functions run plain Node web servers on Lambda, no OpenNext needed — samgoodwin89 · 2026-10-27
- Alchemy Cuts Cloudflare Bindings CPU Time by Up to 166× Via Profiling — samgoodwin89 · 2026-10-27