ChatGPT o1 now has image input capabilities. Which means you can upload a flowchart and ask for high-quality code to build it. *code isn't perfect but better than 4o
LLMS
-
Reasoning Models Perform Better With Simple Prompts
By
–
As OpenAI writes in their documentation (https://platform.openai.com/docs/guides/reasoning#advice-on-prompting), these models perform best with simple prompts. "Advanced" techniques, like few-shot prompting or instructing the model to "think step by step," may not enhance performance and can sometimes hinder it.
-
Internal Reasoning Makes Step-by-Step Prompts Unnecessary
By
–
• Avoid chain-of-thought prompts: Since this model performs reasoning internally, prompting them to "think step by step" or "explain your reasoning" is unnecessary.
-
Keep AI Prompts Simple and Direct for Better Results
By
–
• Keep prompts simple and direct: This model better understands and responds to brief, straightforward instructions without extensive guidance.
-
OpenAI Announces o1: Key Tips for Users
By
–
Since lots of you will be using o1 in the next few days, thanks to today's @OpenAI announcements, remember to…
-
PaliGemma 2 Running 100% Local On-Device with MLX
By
–
PaliGemma 2 running 100% local, on-device, powered by MLX 🔥
— Vaibhav (VB) Srivastav (@reach_vb) 5 décembre 2024
pic.twitter.com/pSHyu9cfgs https://t.co/xB17RN3M4yPaliGemma 2 running 100% local, on-device, powered by MLX
-
LLM comparison: which models solved the contrived problem?
By
–
Contrived problem, I know, but: ChatGPT 4o, o1-mini, and Claude 3.5 Sonnet all get this wrong — 0 out of 3 each ChatGPT o1 gets it right 3 out of 3
-
SmallCon 2024: Fine-tuning SLMs and Real-world AI Use Cases
By
–
Last chance to sign up for #SmallCon (Dec 11)! Hear from leaders at @nubank
, Harvey AI, @rungalileo
, @DoorDash
, & @Convirza on: When to fine-tune #SLMs Real-world use cases Avoiding production pitfalls Register now: https://
pbase.ai/4ij1EZT -
Abacus ChatLLM Teams: Multi-Modal AI Agent Platform with Multiple LLMs
By
–
Get @AbacusAI ChatLLM Teams: https://t.co/73VwIlbKFB <=> able to summarize, to build #AI agents from, or to chat with multi-modal data sources (PDF, text, images…), using any #LLMs :: Llama-3 on Grok, Claude Opus, Gemini, GPTs, or Abacus Smaug (an incredible open-source model). pic.twitter.com/F8drm4R6kb
— Kirk Borne (@KirkDBorne) 5 décembre 2024Get @AbacusAI ChatLLM Teams: https://
chatllm.abacus.ai/?token=kirk able to summarize, to build #AI agents from, or to chat with multi-modal data sources (PDF, text, images…), using any #LLMs :: Llama-3 on Grok, Claude Opus, Gemini, GPTs, or Abacus Smaug (an incredible open-source model).