@Uber good luck bro 73 minutes kebab walk I guess RIP my hot food Can I get a refund at this point ?

Is this normal im Japan ? #tokyo #japan

LTC little undervalued

I don’t understand these price and valuation gaps. They make no sense to me. Frontier is still Claude Fable 5.1 and GPT-6 Astra. A nose, not a generation. Under them: Qwen3.8 Max at $2/$6 per million tokens, open weights. Kimi K3 at $3/$15 per million tokens, open weights. GLM-5.3 at $1.40/$4.40 per million tokens.What most people should use is cheaper still. DeepSeek V4.1 Flash at $0.15/$0.60 per million tokens. Qwen3.8 Flash at $0.15/$0.47 per million tokens. GLM-5.3 Flash at $0.15/$0.50 per million tokens.Gemini Flash, Grok 4.6, Muse Spark and MiniMax M3 sit in that same band. The $10/$50 per million SKU is the outlier.

For almost all current everyday work, Flash is enough. Writing, search, office, homework, most coding, most agents. Use DeepSeek V4.1 Flash, Qwen3.8 Flash, or GLM-5.3 / GLM-5.3 Flash. Keep Claude or GPT-6 for the last 3%. Anthropic is about $965B on about $65B ARR. OpenAI is about $852B on about $40B ARR. Moonshot is $35B to $50B. Chinese labs are not cheaper companies. They are cheaper intelligence. If most real work no longer needs the luxury model, those valuations are pricing a monopoly the market already broke.