GPT-4.5 is the best model anywhere. Talk to it long enough and you will agree. Fuck the benchmarks.
LLMS
-
LLMs Easily Influenced by Point-Based Games Through Reinforcement Learning
By
–
Yes, quite well in fact. LLMs are very easily influenced by point-based games. It's an artifact of the reinforcement learning process
-
Point System Technique to Ensure AI Task Completion
By
–
Bookmark this: If Claude 3.7 or any other AI isn't following all of your instructions, slap its wrist with a point system: "You currently have 20 points. You always perform your task in its entirety, step-by-step, to the best of your ability (which is extremely high), in its
-
Prompt Engineering Tips for Claude 3.7 Model Optimization
By
–
Appending “only do this, nothing more” or “make these edits and do nothing else” type phrases to the end of Claude 3.7 thread chats actually makes a difference. It’s almost TOO proactive without it and breaks code or alters writing too much. HT @mattshumer_
-
Turbocharge Llama AI Agents with Fast Inference and Multi-Agent Orchestration
By
–
Join our webinar with @joaomdmoura / @crewAIInc
, @ShadiCopty / @AIatMeta & @GrahamRSteele / Groq. They'll show you how to turbocharge Llama-based AI agents with ultra-fast inference, scalable conversational AI & multi-agent orchestration. -
Building Full Stack Apps with LLMs in 6-8 Weeks
By
–
From the creator: "I can't believe I built a complex full stack application. All within 6-8 weeks. It has allowed me to teach others skills which will help them 10x their productivity with LLMs." —Abhijat
-
Learn Python with LLMs: Build AI Applications Effectively
By
–
Hard decision to make… Please don't be that guy. Leverage LLMs the right way. We have a new course teaching Python leveraging It gives you just enough Python to start creating AI applications, without spending endless weeks in tutorials. Learn Python by doing! And if you
-
GPT-4.5 Now Live on Perplexity for Pro Users
By
–
GPT-4.5 is now live on Perplexity! Available to all Pro users for up to 10 questions daily
-
CodeLLM Major Upgrade: Fast Autocomplete, Thinking Agents, SOTA Performance
By
–
CodeLLM got a huge upgrade yesterday
— Abacus.AI (@abacusai) 1 mars 2025
– blazingly fast autocomplete
– sonnet thinking for agents
– SOTA coding agent
– multiple UX improvement and bug fixes pic.twitter.com/TSzUEE3kf6CodeLLM got a huge upgrade yesterday – blazingly fast autocomplete – sonnet thinking for agents – SOTA coding agent – multiple UX improvement and bug fixes
