![Pratik Bhavsar Profile](https://pbs.twimg.com/profile_images/1771187314879803393/Je_AdpM9_x96.jpg)
Pratik Bhavsar
@nlpguy_
Followers
3K
Following
5K
Statuses
5K
└► High performance GenAI / Founder @maxpool_ & https://t.co/xYSajPp15C / @iitbombay / https://t.co/90xzcVM69c
Ahmedabad, India
Joined December 2013
@ravithejads @rasbt Everyone should build their own datasets and models..because at the end you grok only what you do.
0
0
1
@averma12 @perplexity_ai hehe.. yes, it is, and they can remove the formatting needed for citations.
0
0
0
RT @ravithejads: 💡 Evaluating Agents 🤔 As I speak with more customers, I have noticed a recurring theme—many are confused with evaluating…
0
1
0
RT @TheTuringPost: A new free eBook: "Mastering AI Agents" from @rungalileo. This ~100-page guide covers: - Agent types and their optimal…
0
14
0
RT @rungalileo: 🔍 Want to build your own AI research agent with @OpenAI's Deep Research? We'll show you how. @nlpguy_ built a comprehensiv…
0
1
0
RT @tuzhaopeng: Are o1-like LLMs thinking deeply enough? Introducing a comprehensive study on the prevalent issue of underthinking in o1-l…
0
103
0
Can anyone from the community answer this? DeepSeek R1 does not support function calling abilities OOTB. Can you share details on the prompt engineering? Is it responsible for calling the tools, or does it handle only specific parts of the full chain?
All Perplexity Pro users now get 500 daily DeepSeek R1 queries (without censorship and prompts not going to China). Free users get 5 daily queries.
0
0
1
@AravSrinivas But DeepSeek R1 does not support function calling abilities OOTB. Can you share details on the prompt engineering? Is it responsible for calling the tools, or does it handle only specific parts of the full chain?
0
0
1
RT @Thom_Wolf: Finally took time to go over Dario's essay on DeepSeek and export control and to be honest it was quite painful to read. And…
0
520
0
RT @AlexGDimakis: What if we had the data that DeepSeek-R1 was post-trained on? We announce Open Thoughts, an effort to create such open…
0
26
0
@abhi1thakur Best wishes for the new adventure! I trained a lot of models with AutoNLP and it supercharged our model building @enterpret_ai.
0
0
1