Viewing profile — fbnbr

idk feels a bit overstated. CUDA’s moat isn’t just writing kernels, it’s the whole ecosystem + hard earned perf intuition. AI helps write code but doesn’t replace that. switching c…

None of that talking about the cost of running such an attack and what models were involved during which phases. Seems like you can use Anthropic now as a proxies bot net

Chat with any AI. Build Your Own AI Agents. Easy to use RAG for Everyone.

Benchmark AI models on standard datasets like FinanceBench and MMLU.

There's some fascinating research relating to this happening at UC Berkeley's Almeida Lab ( https://nature.berkeley.edu/almeidalab/ ) on grapevine diseases, particularly relevant t…

Given the recent Supreme Court decision to overturn the Chevron deference doctrine, the responsibility for interpreting legal gaps now firmly rests with the courts, pushing Congres…

This RouteLLM framework sounds really promising, especially for cost optimization. It reminds me of the KNN-router project ([ https://github.com/pulzeai-oss/knn-router](https://git…

Which Gemini was used is important too btw. - just tried Gemini-1.5-pro and it was working just fine. So I really think the newer versions of LLMs are able to catch this.

That's an interesting prompt I tried on our Pulze.ai platform Spaces and we nailed it with automatically choosing the right model for this type of question gpt-4-turbo: "Yes, you c…

That’s an interesting use case, we thought of a composition of expert router with specialized models like code-generation models to handle coding tasks. Currently we encourage User…

The title is misleading. I don’t see Mesos mentioned ones in the article. I got so excited about reading for Mesos helping in the multi cloud world, potentially as the hypervisor f…

This just becomes interesting if electricity can be produced from reflected photons by the moon such as at night energy production is possible. Other than that I believe in fusion …

“What are LLMs but humans with extreme amnesia and no central coherence?” LLMs are nothing alike to compare them with anything close to what’s going on in a brain. It’s all probabi…

This is way better than google glasses and less nerdy as apples vision

I noticed this really depends on how much data you provide but it’s still cheaper than doing llama2 yourself

Why would you need it as cpp? Do you mean just llama or specifically a version small enough running on edge?

I think the comments on great api design got me thinking of a world in which you can have multiple of these frameworks orchestrate together. I could see use in adding this to a pla…

I have a few of them and it’s really sad to see how Philips is making those harder and harder to use over the years

Open source AI - OSAI - does not necessarily mean it is open. Take the latest OSAI models and try to run them. If you really want to support the idea of open source you’d need to b…

I think smaller expert models will dominate the majority of applications. there is an optimum and fine balance to strike when it comes to size and usability. There will be many mec…