Deepseek beats Qwen in Design with 10% of the token cost
Model tested - Models : DeepSeek V4.1 Flash and Qwen3.8-Flash, head to head both run through omp / opencode-go - Use case : coding - a real feature on my hackat
DeepSeek's open 485B-parameter Flash variant of V4.1.
“I've tested a lot of smaller/cheaper models these past couple of years - this is the first model I can actually hand a detailed spec and have it execute reliably. Building a full Android app in an afternoon with a total bill of $1.20 is something I have never seen any model do!”
“I'm not sure I gave Deepseek-v4.1-Flash a fair shot, but in my limited testing, it spends 20 minutes thinking about relatively basic prompts, so even if it's cheaper than GLM, the time saved by GLM-5.3's much more reasonable thinking time/tokens means you end up ahead if you value your own time.”
“I wouldn't even want to use deepseek-v4.1-flash if it were handed to me. It's fast, sure, but it's brute-force and mindless - the quality isn't high.”
Oui — DeepSeek V4.1 Flash a un score worth-it de 60%, basé sur 364 votes de la communauté.
DeepSeek V4.1 Flash est un modèle de Chat. Way too fast
$0.15/$0.60 per 1M
Model tested - Models : DeepSeek V4.1 Flash and Qwen3.8-Flash, head to head both run through omp / opencode-go - Use case : coding - a real feature on my hackat
Models tested - Model A : Qwen3.8 Flash qwen3.8-flash , run through oh-my-pi omp , thinking level medium - Model B : DeepSeek V4.1 Flash deepseek-v4.1-flash , r