LLM Maven

5.2K posts

LLM Maven banner
LLM Maven

LLM Maven

@GeeFingBeeMan

constantly finding the edge

3rd stone from the sun Katılım Ağustos 2023
2.5K Takip Edilen810 Takipçiler
Sabitlenmiş Tweet
LLM Maven
LLM Maven@GeeFingBeeMan·
no more X bookmarks! This is the Hermes use case unlock. Tell Hermes to create a skill. I post X tweets to Hermes via WhatsApp and it can research it, remember it or go to github to install it.
LLM Maven tweet media
English
0
0
0
162
LLM Maven
LLM Maven@GeeFingBeeMan·
@DrSamuelBHume we are on the verge of getting updated pancreatic cancer results (humans, not rats) from Alpha Tau Medical. if the results are anywhere close to what they previously showed at ASCO, then it will be time for all decent people to march on the FDA to demand immediate cert.
English
0
0
0
20
Samuel Hume
Samuel Hume@DrSamuelBHume·
At the start of this year I wrote about the 5 most important clinical trials to watch 1. RAS inhibition in pancreatic cancer 2. Targeting inflammation to cut cardiovascular risk 3. Targeting lipoprotein(a) to cut cardiovascular risk 4. Tirzepatide vs. Retatrutide in obesity 5. A next-generation amyloid antibody Read it here! substack.com/home/post/p-18…
Samuel Hume tweet media
English
5
32
168
12.9K
LLM Maven
LLM Maven@GeeFingBeeMan·
@exitFiapp @FinnStockinger lol. No! Bad bot! $DRTS is priced 3x higher and theyre announcing Pancreatic cancer results and targeting Japan where they already have PMDA/Shonin approval for Head & Neck cancer.
English
1
0
1
36
Finn Stockinger
Finn Stockinger@FinnStockinger·
The "Multibagger" List is in. Now, how do we find the REAL winner? You guys delivered some incredible alpha. Here’s a snapshot of the tickers mentioned by the community so far: $NIXX | $FPS | $ATOM | $UAN | $AEHR | $LWLG | $M7U | $LPKF | $IQE | $KULR | $DRTS | $OSS | $SIDU | $SLOIF But let’s be real: Buying everything on this list is a recipe for mediocrity. Diversification is a hedge against ignorance, but concentration is how you build real wealth. If you want a 10x return, you can't just follow the crowd. You need a filter. So, I’m asking you: How do YOU pick the "The One"? What is your personal filter for a 10x play? Or do you have another "dark horse" ticker that belongs on this list? The discussion continues below. Share your process. 👇
Finn Stockinger@FinnStockinger

Time to find the NEXT 10x together. 🚀 Yesterday, I shared my conviction plays: $SATL, $KULR, and $IQE. My strategy is simple: look where the crowd isn’t looking yet. Now, I want to hear from YOU. We have an incredible community of alpha seekers here – let’s use it. Share this post with the entire community, let's do this together! Rules of the game: 1️⃣ Drop your pick for the next multibagger in the comments. 2️⃣ Use $TICKER + a short brief (Why is this an asymmetric play?). 3️⃣ You have 24 hours. What’s next?I will gather all the data, analyze the sentiment, and publish a "Community Alpha Report." The most-voted tickers will get their own deep-dive analysis posts from me. Let’s find the next "hidden gem" before the market wakes up. 💎 Share post to whole communit Start below! 👇

English
11
0
28
4.5K
LLM Maven
LLM Maven@GeeFingBeeMan·
@DanielVarab @Teknium i stopped bookmarking things on X. i just send the X post to Hermes via WhatsApp and it knows to read the article, find and install the repo, or offer me an explanation.
English
1
0
4
53
Daniel Varab
Daniel Varab@DanielVarab·
@Teknium im super curious. what are y'all using hermes agent for? i'm intrigued but can't wrap my head around use cases.
English
4
0
5
522
Teknium (e/λ)
Teknium (e/λ)@Teknium·
Alright, the big Question: Which OS do you run Hermes Agent on?
English
47
5
70
8.7K
LLM Maven
LLM Maven@GeeFingBeeMan·
@DrSamuelBHume we are on the verge of getting updated pancreatic cancer results (humans, not rats) from Alpha Tau Medical. if the results are anywhere close to what they previously showed at ASCO, then it will be time for all decent people to march on the FDA to demand immediate cert.
English
0
0
3
293
LLM Maven
LLM Maven@GeeFingBeeMan·
@NousResearch finally! Ive been waiting hours. but seriously - is litellm really truly fully gone?
English
0
0
2
421
Nous Research
Nous Research@NousResearch·
Hermes Agent v0.5.0 is out:
Nous Research tweet media
English
65
92
1.4K
91.9K
Brando
Brando@APIdeclare·
@GeeFingBeeMan @LottoLabs Even on m4 32gb not much better. I think a 14 or 20mb model would be a good sweet spot for Mac users. Funny thing is we’ll look back in 5 years and laugh at how slow these things were
English
1
0
3
21
Lotto
Lotto@LottoLabs·
We need a really good auditing skill for Hermes’ that alone would make 9b/27b more robust
English
6
1
74
4.9K
Ahmad
Ahmad@TheAhmadOsman·
@thbrdy Will have everyone’s requests approved today, thank you for your patience 🙏
English
4
0
11
553
Ahmad
Ahmad@TheAhmadOsman·
I am happy to announce 2 new additions to the x/LocalLLaMA community mod team - 0xSero - sudoingX Hoping this helps streamline growth of the local inference community and deliver grounded, real content over hype Find the community spotlighted at the top of our profiles to join
Ahmad tweet mediaAhmad tweet media
English
28
14
454
25.4K
LLM Maven
LLM Maven@GeeFingBeeMan·
@Abombination81 Coding with Codex and Claude Code. zero clue what I'm doing beyond that.
English
0
0
0
90
Abomination
Abomination@Abombination81·
Ok Need a question answered. Seeing if I can build something for you Guys/Gals/Whatever's. This is very important you answer, for your own benefit. Does everyone here use AI to code? if so which one. If you code, what are you using, python? rust? etc Do you know how to use a json? cli? mcp? md files? I need to figure out how to build things... like back testers, traders, etc for you guys. But need to understand how others are doing things. My way is just one of many. Just put what you know below, IM going to scrape it all and try and make it all work. So if you don't reply, and shit doesn't work... well go punch yourself in your no no spot because it's your own fault.
English
57
2
38
3.6K
LLM Maven retweetledi
Alpha Tau Insight
Alpha Tau Insight@AlphaTauInsight·
JUST IN: Alpha Tau (NASDAQ: $DRTS) to advance Pancreatic Cancer trial in Japan. Following the marketing approval for their Alpha DaRT treatment in H&N Cancer, and the amazing results for Pancreatic Cancer in other countries, DRTS is set to start the trial via their Japanese partner Hekabio.
Alpha Tau Insight tweet media
English
0
4
8
216
LLM Maven
LLM Maven@GeeFingBeeMan·
@ComfyUI @Salvador1993490 I am sad. I went to the ComfyUI web page and the link for "Want to use it locally" was there and then gone so quickly and I can't figure out how to get back to it. So much sadness could be wiped away so easily if Comfy cares.
English
0
0
0
15
LLM Maven
LLM Maven@GeeFingBeeMan·
@sudoingX me too. but i may never know who Alex is...
English
0
0
0
6
Sudo su
Sudo su@sudoingX·
11.7K two days ago. 13.3K today. 1,600 stars in less then 48 hours because you all showed up for majestic Hermes agent . but the hype settles and people forget. the devs don't forget. they're still merging 300+ PRs, still fixing bugs, still shipping parsers for every model you run locally. this isn't a corporation with unlimited funding. this is a small open source team building the tool that keeps your data on your machine and your thinking off someone else's server. they're not asking you for millions in funding. they're asking for adoption, genuine feedback, and a star on a repo. that's it. openclaw has corporate backing and charges you per token. hermes agent is fully open, fully local focused, fully free. 13.3K stars is momentum. 100K is impact. it changes how the industry treats open source AI agents. if you're using it, star it. if you know someone still paying per token for agent sessions, send them the repo. takes 2 seconds. github.com/NousResearch/h…
Sudo su tweet media
Sudo su@sudoingX

star counts matter more than you think. hermes agent is a small team shipping against corporations with unlimited budgets. every star is a signal to developers, contributors, and companies that open source local AI has real demand behind it. NousResearch is open source from the models to the harness. no hidden weights, no gated access, no per token fees. that level of transparency deserves community behind it. 10K stars is recognition. 100K stars is impact. it changes how the industry treats open source AI agents. it brings more contributors, more parsers, more platform support, more models tested. heralds, go star the repo. share it. tell someone who's still paying per token that there's another way. together we make 100K happen.

English
18
19
312
14.1K
LLM Maven
LLM Maven@GeeFingBeeMan·
@ChikoosJourney @LottoLabs more context is good. my point is that youre not going to see sufficient ram reduction such that you can run 27B on a 24GB Mac. the amount of context becomes irrelevant if it kernel panics
English
1
0
2
50
Lotto
Lotto@LottoLabs·
Qwen 3.5 27b is the sweet spot because it’s not a dirty moe
English
23
13
418
37.1K
LLM Maven
LLM Maven@GeeFingBeeMan·
until i get TurboQuant, I'm not able to use Hermes to code what I really want... Claude Cowork but with local LLM ability if you dont want this, you never used it if you do want this and can code it, ill use it
English
0
0
0
224