DeepSeek's Extreme Cost-Performance: Magic or Subsidies?
Fuckinglivemealone · reddit · 2026-07-18
While browsing the Artificial Analysis LLM leaderboard, the author was astounded by DeepSeek's cost-performance ratio. Although the industry universally recognizes DeepSeek's dominance in price-to-performance, its exceptionally high scores are still hard to believe. The author wonders whether this is simply the result of API market subsidies or if DeepSeek has genuinely achieved breakthrough progress in underlying model optimization.
More from Models
- 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
- OpenAI rolls out voice in GPT-Live, but the UI obscures search and reasoning — Graham_dePenros · 2026-07-22
- Moonshot’s Kimi K3 sets a new open-weights ECI record at 156 — scaling01 · 2026-07-22