Hi Friends,

Even as I launch this today ( my 80th Birthday ), I realize that there is yet so much to say and do. There is just no time to look back, no time to wonder,"Will anyone read these pages?"

With regards,
Hemen Parekh
27 June 2013

Now as I approach my 90th birthday ( 27 June 2023 ) , I invite you to visit my Digital Avatar ( www.hemenparekh.ai ) – and continue chatting with me , even when I am no more here physically

Monday, 22 July 2024

SIMULTANEOUS POWERING WITH SEVERAL LLMs

 Kishan ,

 

I came across following news in today’s Eco times , where the second para reminded of your following Whaysapp message two days ago :

 

“  Yes, we can definitely insert an image on that page. As for GPT-4o-mini, I’ve refactored the backend API code to

 

support plug-and-play functionality with various models, including LLAMA and AnthropicThis allows us to easily switch

 

to any other model at any stage  “

 

 

I suppose what you have done is to “ make a provision “ for SWITCHING ( from one LLM to another ), given a short notice in future, if required

 

 

Whereas the new report hints at the possibility of getting answers from SEVERAL LLMs , “ Simultaneously “ , against a query

 

 

Am I right in my interpretation ?

 

 

Would it help  www.My-Teacher.in > Mock Test feature , if we could follow that SIMULTANEOUS mode ?

 

 

hcp

 

 

 

 

 

 

 

 

 

 

OpenAI’s cheaper mini brings a mega issue for LLMs   …  Eco Times … 23 July    

 

 

 

 

Extract :

 


OpenAI
's pricing for its first-ever small language model GPT4o Mini--which is 60% cheaper than GPT 3.5

 

Turbo and 40% cheaper than Google's Gemini 1.5 Flash--has raised the competition bar among Indian makers of LLMs (large language models) who are building small and concise use-cases for organisations.

 

 

 

Executives of startups said OpenAI's pricing could bring massive cost savings for scaling up simple use-cases,

 

such as multilingual chatbots where a single query ranges between 15-25 tokens. OpenAI has also allowed

 

calling multiple APIs (application programming interface), which means a user can pass contextual data from

 

multiple sources at the same time to answer a complex query.

 

 

Multilingual translation at scale is useful in education, book publishing and generating synthetic data

 

for

 

machine learning purposes. Large enterprises can use AI to analyse large volumes of unstructured data, such

 

as social media conversations, to distil consumer insights,” he said.

No comments:

Post a Comment