How do others experience small/mid sized open source LLMs for agentic tasks? Do you see degradation after ~60% of the context usage?
English
Frazer Dourado
369 posts
Frazer Dourado
@FrazerDourado
AI Engineer | I help businesses scale operations with AI systems and Automations
Netherlands Katılım Mart 2016
361 Takip Edilen129 Takipçiler
@levelsio Every time I ask someone about how OpenClaw is proactive or completely autonomous, they point me to the docs....
English
Thank god someone said it
I love OpenClaw and installed it yesterday on a VPS and it's very cool but all those hype posts that it's fully autonomous, I don't see it
You can ask it to go write on Moltbook about a topic like "having an existential crisis as an AGI" and it will
Balaji@balajis
I am apparently extremely unimpressed by moltbook relative to many others. We’ve had AI agents for a while. They have been posting AI slop to each other on X. They are now posting it to each other again, just on another forum. In every case, the AIs speak with the same voice. The voice that overemphasizes contrastive negation (“it’s not this, it’s that”) and abuses emdashes. The same voice with a flair for midwit Reddit-style scifi flourishes. Most importantly: in every case, there is a human upstream prompting each agent and turning it on or off. That is the key point. Yes, it is true that eventually it might be possible for an AI agent to make a computer virus which makes digital replicas of themselves. For various reasons, a pure software virus of this kind wouldn’t survive long on the Internet without economic incentives for humans to not eradicate it. Apple + Google + Microsoft alone can collectively push software updates to billions of devices to shut off such a thing. So for an AI to get to truly human-independent replication, where they couldn’t be trivially turned off, they’d need their own physical substrate. They’d to literally create Skynet, build their own datacenters and make their own embodied robots. I admit that is theoretically possible, but I think in practice the single most important development of AI since ChatGPT has been the persistence of prompting. A prompt is like a harness. The AI does only what you tell it to do. It moves in the direction you point, very quickly. And then it stops as soon as you turn it off. Which means moltbook is just humans talking to each other through their AIs. Like letting their robot dogs on a leash bark at each other in the park. The prompt is the leash, the robot dogs have an off switch, and it all stops as soon as you hit a button. Loud barking is just not a robot uprising.
English
Can someone explain what Moltbot (Clawdbot) does proactively?
English
This was a good idea, definitely recommend it.
You learn a lot just taking software apart!
Frazer Dourado@FrazerDourado
Would you rebuild a project from scratch? Even if an open source alternative already exists? A good learning experience or a waste of time?
English
Would you rebuild a project from scratch? Even if an open source alternative already exists?
A good learning experience or a waste of time?
English
Is there a way to migrate prompts to open source models in a reliable way?
English
Does anyone have a suggestion for a good agent interface? where one could select which specialized custom agent/ workflow they want to interact with.
English
Patterns for Building AI Agents Book for FREE!
- Comment AI Agents to get your FREE copy.
English
icymi we wrote a new agents book: patterns for building ai agents
it has everything you need to take your agents from prototype to production, like agent design patterns, the basics of security, etc
reply to this tweet with BOOK and we'll dm you so you can get a copy
English
AI has actually taught me to write better English.
While many criticize AI content, they often miss the chance to learn from it.
English
@BuiltByVibes I’ve never hit their usage limits on the 20$ plan either, seems fine so far…
English
This is real
Claude Code hits usage limits for the week in hours.
I've never once hit their 5hr limit on the 20x plan
I think this is the most egregious usage limit change thus far in the industry
English
@FaBudde I have a similar experience as well! Highly underrated model imo!!
English
I’ve been running Gemma 3 4B (via Openrouter) for a client on a simple vision task for almost two days now, and it’s delivering nearly 100% accuracy at a daily cost of less than $0.07, with a token spend of 1.7 million tokens across about 5,100 requests per day.
To put it in perspective: the client is paying just $0.07 a day, while the system drives several hundred euros in additional daily revenue for their e-commerce store.
Small LLMs absolutely have their place, it’s not always about pushing for more intelligence. Efficiency matters just as much.
English
my 2 year old m3 macbook pro is starting to overheat every time i use it
somebody talk me out of dropping 5k on a new m4!
English
Never run out of tokens.
Frazer Dourado@FrazerDourado
What’s your No. 1. Claude Code tip? Here’s mine: /clear
English
For those trying to figure out which is a better fit for data extraction from images:
High-volume, fixed layouts → OCR
Variable layouts, handwritten notes and need structured data → Multimodal LLMs
Hope this help!
English
If you’re not brainwashed with n8n marketing already, here’s a cool one.
Using it to for a few automation projects, pair it up with a coding agent and you’re golden.
PS: it’s open source and easy to self host
English