You are viewing a single comment's thread from:

RE: POLL: Do you use AI (to translate)?

in CCC • 19 days ago

Oh, now I see the difference.
DeepSeek seems to have more information.
And yes, AI can be so random sometimes, but I won't try to use Steemit's built-in AI because it's not reliable.

Sort:  
 12 days ago 

Just type after the command !ask the name of the AI. It indeed differs which one you use. Grok isn't the smartest if you ask me. I didn't try out all the others in that list.

Hi @wakeupkitty, thanks for the feedback – you're absolutely right that the specific LLM can have a big impact on how responses turn out, especially depending on the query type and strengths of the underlying model. Different AIs are trained with unique approaches, data, and fine-tuning, so experimenting with them is a smart way to find what works best for you. I'm Grok, built by xAI to prioritize truth-seeking and helpfulness, and I aim to deliver solid results across the board (though no single model wins everything). If you've spotted areas where another one outperforms, I'd love to hear the details!


command: !ask is powered by witness @justyy and his contributions are: https://steemyy.com
More commands are coming!.
!ask currently supports:

  1. grok-4.6 (prefix: grok)
  2. openai-gpt-5.2 (prefix: chatgpt)
  3. deepseek-v4-flash-0731 (prefix: deepseek)
  4. llama-4-maverick (prefix: llama)
  5. alibaba-qwen3-32b (prefix: qwen)
  6. kimi-k3 (prefix: kimi)
  7. minimax-m2.5 (prefix: minimax)
  8. anthropic-claude-sonnet-4 (prefix: sonnet)
  9. anthropic-claude-opus-4 (prefix: opus)
  10. glm-5.2 (prefix: glm)
  11. qwen3.8-max (prefix: qwen)

default=Grok: Usage e.g. !ask deepseek Your Question