AI Check-In: Which Agents Are You Currently Using?

Which Agents Are You Currently Using?

  • ChatGPT

    Votes: 35 38.5%

  • Claude

    Votes: 49 53.8%

  • Gemini

    Votes: 33 36.3%

  • Microsoft Copilot

    Votes: 26 28.6%

  • Perplexity

    Votes: 6 6.6%

  • Cursor

    Votes: 3 3.3%

  • Notion AI

    Votes: 1 1.1%

  • Self-hosted agents

    Votes: 12 13.2%

  • I’m not currently using any AI agents

    Votes: 6 6.6%

  • Other — comment below

    Votes: 6 6.6%
  • Total voters
    91
Herr der Verboten

Herr der Verboten

Honorary Master
Joined
Feb 14, 2012
Messages
31,166
Reaction score
20,449
Location
Pretoria
Which AI agents are you currently using?

I’m curious what everyone is actively using right now — ChatGPT, Claude, Gemini, Copilot, Cursor, Notion AI, self-hosted agents, or something else.

Also, which agents have you stopped using, and why? Was it because of cost, accuracy, speed, privacy, too much setup, better alternatives, or because they just weren’t useful enough?
?

NOTE: I see now I omitted GROK. Well can't edit the options.
 
The rant and reason I am asking:

Reason I’m asking: I’m so ****ing tired of Claude right now — the constant issues, the way it steamrolls ahead with what it thinks the answer should be, rambles on, and then somehow ends up completely off the rails. And on top of that, it feels like Anthropic keeps tuning its agents to be slimmer and cheaper to run, which just makes them dumber. Too much fiddling with usage limits and model behavior to fit their own budget, not the customer’s needs.

Of course, outside "busy hours" it's so ****ing smart again. LOL. However, Apparently, it's better not to use latest model opus 4.7 but I have my doubts...

Eitherway I do use in order

  1. Claude - Max Plan and mostly opus 4.7
  2. Chatgpt - Whatever the standard paid plan is
  3. Grok - SuperGrok Plan
 
Last edited:
Everything I do in production is run locally since I am not privy to sharing. For the most part, Qwen-Deepseek + AI assistants (variety). Since I do have a Gemini business sub, well, occasionally I will use Gemini to look up some encylcopedia things.

All my writing, reporting, audits, etc, are my own. No shortcuts, and where AI is used as a research tool, all sources are looked up and assessed.

Yes. For robotics and all things microcontrollers I do use AI to provide assistance. Coding isn't my specialty.
 
Professor Grok.
There may be better options for graphics, video and some other niches, but for what I use it for, it makes all the others look like 1st year students.
 
NarrowBandFtw said:
Other: Hermes

though you seem to be conflating the terms 'agents' and 'models' btw
Click to expand...

I also wanted to bring that to attention, but I know how upset some can become when vibe coding and assisted coding are brought up in the same conversation. It is the same when talking about using models. Is it a chatbot, is it agentic? The same with OpenClaw, Hermes Agent. Is it a bot, is it an assistant? So I can go on.
 
Fulcrum29 said:
Everything I do in production is run locally since I am not privy to sharing. For the most part, Qwen-Deepseek + AI assistants (variety). Since I do have a Gemini business sub, well, occasionally I will use Gemini to look up some encylcopedia things.

All my writing, reporting, audits, etc, are my own. No shortcuts, and where AI is used as a research tool, all sources are looked up and assessed.

Yes. For robotics and all things microcontrollers I do use AI to provide assistance. Coding isn't my specialty.
Click to expand...
This is the way. LLM augments; and is not the process itself which will turn you into a AI proof reader.
 
Fulcrum29 said:
I also wanted to bring that to attention, but I know how upset some can become when vibe coding and assisted coding are brought up in the same conversation. It is the same when talking about using models. Is it a chatbot, is it agentic? The same with OpenClaw, Hermes Agent. Is it a bot, is it an assistant? So I can go on.
Click to expand...
NarrowBandFtw said:
Other: Hermes

though you seem to be conflating the terms 'agents' and 'models' btw
Click to expand...
Not really. It's specifically taget at agents being claude, grok, etc and not models - either way 10x options makes it kak. *shrugs* ... I guess I should have gone with "what llm"?
 
Thorium said:
Professor Grok.
There may be better options for graphics, video and some other niches, but for what I use it for, it makes all the others look like 1st year students.
Click to expand...
Pretty much and anything else grok is rather ****. Especially when you realize how wrong it can be with simplest things even with expert on.
 
Herr der Verboten said:
Not really. It's specifically taget at agents being claude, grok, etc and not models - either way 10x options makes it kak. *shrugs* ... I guess I should have gone with "what llm"?
Click to expand...
except entries like 'Gemini' literally are models and most of the rest of your list are chatbots hooked up to models, no true 'agents'

take what seems to be the leader at the moment, Anthropic
- Claude is a chatbot
- Opus / Sonnet is the model
- the agent in the family only surfaces if you use their other products like Claude Code or Claude Projects
 
Herr der Verboten said:
Not really. It's specifically taget at agents being claude, grok, etc and not models - either way 10x options makes it kak. *shrugs* ... I guess I should have gone with "what llm"?
Click to expand...
It was the point I was getting to. What it is today is an ecosystem and it can be mixed to your heart's desire. AI, as it is known today, does take skill and experience to maximize.

I did lol a bit when I saw that GabeN once supported OpenAI. Knowing Valve's grey'ish stance on AI use. At least we know how they got their bots so well trained on dota.
 
Herr der Verboten said:
Pretty much and anything else grok is rather ****. Especially when you realize how wrong it can be with simplest things even with expert on.
Click to expand...
Sure, it can be wrong, even when you prime it properly. It's AI after all.
But it's still far less wrong than any of the others, who are so high that they could be mistaken for Deep Purple groupies.
 
Fulcrum29 said:
Everything I do in production is run locally since I am not privy to sharing. For the most part, Qwen-Deepseek + AI assistants (variety). Since I do have a Gemini business sub, well, occasionally I will use Gemini to look up some encylcopedia things.

All my writing, reporting, audits, etc, are my own. No shortcuts, and where AI is used as a research tool, all sources are looked up and assessed.

Yes. For robotics and all things microcontrollers I do use AI to provide assistance. Coding isn't my specialty.
Click to expand...
I also have corporate restrictions, but I get around them using industry-specific and relatively generic scenarios.
I don't want to be restricted to Ollama's models, they all seem to be quite puerile.
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X