Izzat Raihan
89 posts
Izzat Raihan
@izzatraihan
Tweeting what I like: tech, AI, travel, good food. Delivering happiness with https://t.co/x8Yk7NR27l
Berlin, Germany Katılım Mayıs 2023
198 Takip Edilen27 Takipçiler
@izzatraihan long story hahaha; I have whole tutorial covering this project youtu.be/yGQb9KkvQ1Q
YouTube
English
supervision just hit 40,000 GitHub stars!
it now powers over 6.5k open-source computer vision projects, including all my demos like basketball AI
link: github.com/roboflow/super…
English
@LilTea_eth @PaulGugAI @HermesAgentTips yees, you can tell Hermes directly. first you need to setup the auth tho. hermes model -> setup deepseek auth. once it’s done, tell hermes to setup the main model to be codex gpt5.5 and then fallback to deepseek. happy to help if you found issues
English
@izzatraihan @PaulGugAI @HermesAgentTips How to set it up, I'll study it, can you tell Hermes directly?
English
be honest... whats the best coding model to use pair with hermes agent?
- Opus 4.8
- GPT-5.5
- DeepSeek V4 Pro
- Opus 4.7
- Kimi K2.6
English
@LilTea_eth @PaulGugAI @HermesAgentTips you can set fallback model, so everytime you got rate limited, it auto switches to the next fallback model
English
@HermesAgentTips GPT 5.5 and DSV4 (if the GPT Plus sub runs out) is the combo I have settled on.
I'd like GPT to be a bit less wordy with responses sometimes, but that's probably something I should tweak in the agent profile..
English
@measure_plan always inspiring. how did you come up with these ideas? crazy
English
sorry can't chat rn i'm grinding guitar fruit ninja until i figure out what the hell is going on with the f chord
AA@measure_plan
i made fruit ninja but you've got a guitar instead of a sword
English
really appreciate the team with smooth resolution! and I got $50 free credits. bullish on browser use 🚀
Browser Use@browser_use
@izzatraihan @gregpr07 We’re looking into this right now! DM us and we’ll help you out as well.
English
Day 11 of #buildinpublic
so excited to have connected with 228 people 🥳
am looking to connect with more #builders, especially:
> AI startups & SaaS
> eCommerce & local businesses
> AI content creators
happy to swap tips on acquiring users, iterating on ideas, and indie hacking in general 🫡🎉
English
Looking to meet more builders on X 🚀
If you love:
🔹 Building SaaS ⚡️✨
🔹 Full-stack dev life 🤓💻
🔹 Shipping in public 🚢🚢
🔹 Learning by doing 🤔🙌🏻
Drop a 👋 and let's build, learn, and grow together
👇
English
@jackprice hey jack, let's connect! cooking something, but not yet live
English
I want to connect with more
founders
builders
vibe coders
AI enthusiasts
UI designers
If you’re someone building with AI right now, drop it in the comments and let's connect
English
@izzatraihan @gregpr07 We’re looking into this right now! DM us and we’ll help you out as well.
English
Introducing: New Browser Infrastructure. ⚡️
Chromium fork. Firecracker fork. Custom Linux kernel.
> 3x cheaper. $0.02/h.
> Unlimited scaling
> Subsecond cold starts
> Most stealth browser infra
Feel the speed. Now live.
Try it now ↓
English
@browser_use hi @gregpr07 I love this new stealth browser. It really worked with my use case. I want to introduce this to my team, but I got an issue with the credits. I was charged by $5 every time I spawn a browser session, although only spent a few cent. anyone from the team can help me?
English
@browser_use I tried it today and it's fantastic! however, my credits got used for about $75, although from analytics it was only about $1.4. anybody can help why it happened? I'm in the free tier, but have credits for pay-as-you-go, and using it via API
English
Gemma 4 12B can now run locally on just 8GB RAM via Dynamic GGUFs.
Google's new model, Gemma 4 12B Unified supports image, audio and 256K context.
You can run and train the model via Unsloth Studio.
GGUF: huggingface.co/unsloth/gemma-…
Guide: unsloth.ai/docs/models/ge…
Google Gemma@googlegemma
Meet Gemma 4 12B! A unified, encoder-free multimodal model designed to bring high-performance intelligence directly to your laptop, and released under an Apache 2.0 license. Bridging the gap between edge efficiency and advanced reasoning. Here is what’s new with Gemma 4 12B: 👇
English
@garrytan the only problem is the frontier labs heavily subsidize the users with their Max/Pro plans
English
Model routing is an important thing
Controversial idea: the frontier labs will want their AI harness to be the moat, but ultimately the best case for consumers is that model capabilities flatten and commodify
Preview of the AI Harness Wars of 2027
Factory@FactoryAI
Introducing model routing to Factory. Factory Router picks the right model for every task, automatically. Maintain frontier performance while cutting costs by 25%.
English