Does LLM Council/Fusion Actually Work?
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Unlock all features
FREE: Get instant access to 10 AI summaries, chats, or transcripts per day.
Related videos
L8 Principal's Agentic Engineering Workflow
Kun Chen
50.6k views
What Does Palantir Actually Want? (Re-upload)
ColdFusion
98.0k views
What Does Palantir Actually Want?
ColdFusion
199.9k views
How does Claude Code *actually* work?
Theo - t3․gg
189.3k views
Sonnet 4.5 Is Here—And It’s a Beast at Coding
Prompt Engineering
52.0k views
Context Engineering is All You NEED!
Prompt Engineering
38.7k views
The Only Embedding Model You Need for RAG
Prompt Engineering
35.2k views
Gemini CLI — Google’s Free Open-Source Coding Agent
Prompt Engineering
56.6k views
AI prompt engineering in 2025: What works and what doesn’t | Sander Schulhoff
Lenny's Podcast
68.3k views
The Secret to Perfect Prompts (Without Prompt Engineering)
Futurepedia
55.4k views
Top Comments (10)
Duuuuuuuuude :| I'm getting a headache so maybe that's why I can't find the link to your LLM Council code like you mentioned again & again & again... What's happened to the channel? It's NOT the same quality as it was in the beginning. Remember when it was exciting & new? Well, that shit's over :p You say your voice is your own, but your workflow is touched by SOOOOO much AI that you'd better proof-read AND proof-watch AND proof-listen to everything before publishing. True story.
Can't wait to pay 3x usage for a tiny improvement in quality. I sure love wasting tokens!
So without Vercel you cannot access multiple providers using a single endpoint?
Been doing this forever. It works but needs the human in the loop
I've tried this out manually when I first started using llms, but by hand, and its true, some are better for certain aspects, one is better at code, one is better at the design, etc, etc. Then you have them each check each others work. I was acting as the human api gateway and orchestrator, but I think most people knew this by human nature, we just didnt do it in the design, we did a 1 vs 1, and agents are just clones of the same provider. People are finally breaking out to multi model interactions. I wrote a quick shell script to try this with lm-studio and 5 models, 1 chairman, 4 diff lm models. The only thing, is I didnt have it anonymized, then I had them work on a project, and they did group vote, and chairman had the final decision. Interested to see how multi model makes into frameworks.
I‘m using open router fusion for this and mostly to review and critique a plan before executing it in a coding agent so it takes the review points and updates the plan if it agrees
Checkout Vercel's AI Gateway: https://vercel.plug.dev/B3rLR9K
Every time I think there is no other way to burn tokens easily something like this appears. A loop is missing, and switching the roles. 🎉 ah, I’ve forgotten to say that at the end there should be a council for the n^2-1 councils.
Underexploited .... correct directionally
you understand that i am seeing this video 3 days after posting and it is so outdated..... LLM change daily...... you can't post a YouTube video...... it is already obsolete....
Unlock the Data Inside
Turn Videos into Knowledge
- Get FREE 10/day: transcripts, summaries, chats
- Chat with videos, export text & PDF
- $1 free API credit for RAG, chatbots & research
Free forever plan • All features unlocked
Top Comments (10)
Duuuuuuuuude :| I'm getting a headache so maybe that's why I can't find the link to your LLM Council code like you mentioned again & again & again... What's happened to the channel? It's NOT the same quality as it was in the beginning. Remember when it was exciting & new? Well, that shit's over :p You say your voice is your own, but your workflow is touched by SOOOOO much AI that you'd better proof-read AND proof-watch AND proof-listen to everything before publishing. True story.
Can't wait to pay 3x usage for a tiny improvement in quality. I sure love wasting tokens!
So without Vercel you cannot access multiple providers using a single endpoint?
Been doing this forever. It works but needs the human in the loop
I've tried this out manually when I first started using llms, but by hand, and its true, some are better for certain aspects, one is better at code, one is better at the design, etc, etc. Then you have them each check each others work. I was acting as the human api gateway and orchestrator, but I think most people knew this by human nature, we just didnt do it in the design, we did a 1 vs 1, and agents are just clones of the same provider. People are finally breaking out to multi model interactions. I wrote a quick shell script to try this with lm-studio and 5 models, 1 chairman, 4 diff lm models. The only thing, is I didnt have it anonymized, then I had them work on a project, and they did group vote, and chairman had the final decision. Interested to see how multi model makes into frameworks.
I‘m using open router fusion for this and mostly to review and critique a plan before executing it in a coding agent so it takes the review points and updates the plan if it agrees
Checkout Vercel's AI Gateway: https://vercel.plug.dev/B3rLR9K
Every time I think there is no other way to burn tokens easily something like this appears. A loop is missing, and switching the roles. 🎉 ah, I’ve forgotten to say that at the end there should be a council for the n^2-1 councils.
Underexploited .... correct directionally
you understand that i am seeing this video 3 days after posting and it is so outdated..... LLM change daily...... you can't post a YouTube video...... it is already obsolete....