DeepSeek Flash's performance is mind-boggling,网友直呼“不合理”
yacineMTB · x · 2026-08-05
yacineMTB tweets showing DeepSeek Flash's performance at the bottom of a chart, saying 'this makes no sense', implying its performance is surprisingly good or confusing.
Related event: DeepSeek V4-Flash Tops OpenRouter, Costs 1% of Rivals(11 posts)→
More from Models
- Frontend Arena Shakeup: Kimi, Qwen, and Claude Dominate Top 3 — YiMaTweets · 2026-08-05
- After 3 Years of Testing AI Sycophancy: Models Still Fail to Be Objective — sdfprwggv · 2026-08-05
- Burning $130K/Day? Unpacking DeepSeek API Token Volumes — teortaxesTex · 2026-08-05
- Mistral AI Introduces Shieldstral for AI Safety and Moderation — tengo_harambe · 2026-08-05
- Testing 12 LLMs on Bug Fixing: The Cheapest Tokens Lead to the Highest Real Cost — deusaquilus · 2026-08-05
- OpenAI's gpt-5.6-luna So Cost-Effective It Constantly Overloads Servers — _lewtun · 2026-08-05