Objective errors

https://t3.chat/settings/models

• Deepseek R1 (OpenRouter) > Show more > it’s → its

• Qwen 2.5 32b > Show more > it’s → its

Inconsistencies/nits

https://t3.chat/settings/models

• Every model show more description ends with a period except Gemini 2.0 Flash

https://t3.chat/chat model selector

• Every model info tooltip ends with not a period except GPT-4o-mini and Llama 3.3 70b

Personal Opinions

https://t3.chat/settings/models

• Deepseek v3 (Fireworks) > Show more > Just…slow → Just… slow

• o3-mini > Show more > I understand “didn't want Deepseek to go too far” can be interpreted like “go far in life” (i.e. to succeed), but the way I read it was like “that joke went too far” (i.e. to do too much). OpenAI releasing a model doesn’t stop Deepseek from doing as much as they want, so it doesn’t really make sense using the second interpretation. I think something more fitting would be to say “OpenAI clearly didn’t want Deepseek to get too far ahead”, or if you want something closer to the first interpretation “OpenAI clearly didn’t want Deepseek to get all the attention”.