lemonish97

68 karma · joined May 18, 2023

submissions comments

lemonish97··on Copilot Cowork: A new way of getting work done

Clearly shows you haven't opened the link?

"Working closely with Anthropic, we have integrated the technology behind Claude Cowork into Microsoft 365 Copilot."

lemonish97··on Phi-4-reasoning-vision and the lessons of training a multimodal reasoning model

Always love a new good SLM. Reasoning + vision under 20B sounds promising, will be testing this out

lemonish97··on OpenAI should build Slack

Honestly prefer teams over slack. Slack is good at text threads, while I'd still choose Teams for calls and meetings.

lemonish97··on Introducing the Developer Knowledge API and MCP Server

So an mcp server just for google specific public docs? Aren't there a dozen of them like Context7 (https://context7.com/) that already do this?

lemonish97··on Microsoft open-sources LiteBox, a security-focused library OS

I know windows 11 is super buggy and riddled with issues (and the copilot mess), but I'm starting to feel there's a weird echo chamber around these forums that don't even bother looking at what the product or repository is, and automatically assume it's bad 'cause it's from Microsoft.

lemonish97··on Windows 11 Outperforming Linux on an Intel Arrow Lake H Laptop

Or maybe it is just better?

lemonish97··on The State of AI Coding Report 2025

Not sure if it's a TPU constraint, but according to this report it seems like the Gemini models have really poor TTFT and tps inference times.

lemonish97··on Fara-7B: An efficient agentic model for computer use

It's great to see how we went from the first iteration of Claude Computer Use, to now being able to run it locally with just 7B params.

lemonish97··on Gemini 2.5 Flash Image

Yes, they mention that the model is aka nano-banana in the blogpost

lemonish97··on Gemma 3 270M: Compact model for hyper-efficient AI

I use PocketPal. Can run any gguf model off hf.

lemonish97··on Gemma 3 270M: Compact model for hyper-efficient AI

Never thought I'd run an LLM released in 2025, on my phone, in full BF16. With ~80tps on an iPhone 16 pro btw.

lemonish97··on Google Is Winning on Every AI Front

Honestly love gemini for zero-shot coding. But for some reason I'd still lean towards GPT4o for just natural conversations and day-to-day queries. Something about 4o's tone and behavior just clicks with me.

lemonish97··on HP Acquires Humane's AI Software

To be fair, HP buying Compaq in the early 2000s was pretty huge, considering they are doing decently well today because of their PC market share.

lemonish97··on NetworkX – Network Analysis in Python

used this throughout my ML on graphs grad course this sem, it’s a really neat library

lemonish97··on LLMs confabulate not hallucinate

Fun fact, it is believed that Andrej Karpathy (OpenAI, ex AI Dir. at Tesla) had coined the word 'hallucination' in an RNN blogpost back in 2015. https://x.com/karpathy/status/1702916988891193460?s=20

lemonish97··on PoW gets his life back after 55 years (2000)

Thought this was some ‘Proof of Work thing’. Def need a break from HN.

← PreviousPage 2 of 2