@Handllex
Account based inFrance
About this account
- Account based in
- France
- Connected via
- France App Store
Account-level information from X, not a live location or the device used for a specific post.
AI Creator & Automation Expert | Higgsfield CPP & Tripo AI Turn X + AI into a $3K/month income stream Open to collaborations 📩 [email protected] or DM
United States
Joined January 2011
- Tweets22.5K
- Following139K
- Followers201K
- Likes33.5K
Pinned Tweet
When a product experience becomes the benchmark others look to, it says a lot.
@wanderboat_ai didn’t just explore a new way to discover places, they helped define it.
Twice in one year, Wanderboat features have shown up in Google Maps.
March: Chat feeds + Vibe Checking
Today: AI Food Ordering, Hotels, Local Events...
Their "biggest transformation in over a decade" is still a demo video (with a "simulated" disclaimer). Ours has been live for 1.5 years.
Feels like Wanderboat has been handing them a free roadmap. And we have more.
GTA VI is getting closer, and AI can already bring that open-world feeling to life. 🎮
Think cinematic streets, immersive characters, dramatic action, and game-like scenes. PixVerse makes it easy to explore these ideas through AI video creation.
Discover what your own AI-generated game world could look like:
app.pixverse.ai/cli/viral-pr…
@PixVerse
Plus, eligible paid-plan members can get FREE & UNLIMITED image generation with GPT-Image 2.5 through Oct 4, 11:59 PM PDT.
Fan-made AI concept. Not actual GTA VI gameplay or official Rockstar Games footage.
#AIVideo #Gaming #GenerativeAI
🤝 Paid partnership · 🤖 Made with AI
AI video generation is getting seriously cheaper 👀
@Flovaai is running a limited-time promotion from Sep 21–Oct 20 with up to 80% off, but the most interesting part is what happens after you claim the deal.
You can lock in the promotional rates for 365 days.
Seedance 2.5 720p starts at just $0.044/sec, while 480p starts at $0.022/sec.
Wan 3.0 Prime starts at $0.022/sec, and Image 2.5 from $0.011/image.
If you’re generating AI content regularly, lower generation costs can mean a lot more room to experiment.
FLOVA SAVINGS MONTHS: Flova × Seedance 2.5
Up to 80% Off. Lock In Your Rate for 365 Days.✨
September 21 – October 20
Seedance 2.5: Up to 80% Off — Enjoy this discounted rate for a full year
- 480p: From $0.022/sec
- 720p: From $0.044/sec
More Exclusive Offers:
Image 2.5: From $0.011/image
Wan 3.0 Prime: From $0.022/sec
FLOVA SAVINGS MONTHS: Flova × Seedance 2.5
Up to 80% Off. Lock In Your Rate for 365 Days.✨
September 21 – October 20
Seedance 2.5: Up to 80% Off — Enjoy this discounted rate for a full year
- 480p: From $0.022/sec
- 720p: From $0.044/sec
More Exclusive Offers:
Image 2.5: From $0.011/image
Wan 3.0 Prime: From $0.022/sec
The 4B active parameter number is what caught my attention here.
K2-Horizon-36B-A4B has 36B total parameters, activates ~4B per token, and reaches 25 on the Artificial Analysis Intelligence Index #4 among 142 comparable models.
That’s a very interesting efficiency curve.
🤝 Paid partnership
K2-Horizon-36B-A4B scores 25 on the Artificial Analysis Intelligence Index, matching models with over 20× the total parameters while using 4B active parameters per token.
These capabilities come from our new architecture MoVA (Mixture-of-Value Attention), which incorporates MoE-based sparsity into the compute of value vectors in multi-head attention.
It opens a second axis for scaling sparsity in an LLM, beyond MoE in the FFN module. Importantly, MoVA enjoys the following advantages:
• Simple and compatible with efficient attention algorithms, such as flash attention, GQA, and sparse attention
• No additional KV cache cost comparing to standard GQA
K2-Horizon-36B-A4B available at: huggingface.co/IFM/K2-Horizo…
Arcane Matrix | AI retweeted
K2-Horizon-36B-A4B scores 25 on the Artificial Analysis Intelligence Index, matching models with over 20× the total parameters while using 4B active parameters per token.
These capabilities come from our new architecture MoVA (Mixture-of-Value Attention), which incorporates MoE-based sparsity into the compute of value vectors in multi-head attention.
It opens a second axis for scaling sparsity in an LLM, beyond MoE in the FFN module. Importantly, MoVA enjoys the following advantages:
• Simple and compatible with efficient attention algorithms, such as flash attention, GQA, and sparse attention
• No additional KV cache cost comparing to standard GQA
K2-Horizon-36B-A4B available at: huggingface.co/IFM/K2-Horizo…
Repurposing the structural magic behind a viral video with millions of views is officially effortless.
Dreamina_ai revamped Canvas with hit Workflows across Films, ads, social media and so on.
Look at the sensory punch of this fine-dining trailer here is how I constructed this project:
Directing your scene sequence through a Workflow on Canvas keeps all your takes organized so you can wire clips straight to the Timeline.
Need cinematic focus shifts? You can also jump into Generate mode to apply camera Skills in a single click.
That’s where AI filmmaking gets serious: Workflows supply the skeleton, Elements anchor subject identity, and Skills deliver precise camera technique.
Stop guessing & build repeatable systems with @Dreamina_ai:
dreamina.capcut.com/ai-tool/…
Comment “Dreamina” and I’ll DM you the free user guide + share the link!
#DreaminaMadeItViral #DreaminaAI #DreaminaAIweb
Arcane Matrix | AI retweeted
GTA VI is getting closer, and AI can already bring that open-world feeling to life. 🎮
Think cinematic streets, immersive characters, dramatic action, and game-like scenes. PixVerse makes it easy to explore these ideas through AI video creation.
Discover what your own AI-generated game world could look like:
app.pixverse.ai/cli/viral-pr…
@PixVerse
Plus, eligible paid-plan members can get FREE & UNLIMITED image generation with GPT-Image 2.5 through Oct 4, 11:59 PM PDT.
Fan-made AI concept. Not actual GTA VI gameplay or official Rockstar Games footage.
#AIVideo #Gaming #GenerativeAI
🤝 Paid partnership · 🤖 Made with AI
Arcane Matrix | AI retweeted
Probé la misma tarea con dos agentes diferentes.
Codex Astra costó $5,30.
OpenCode + DeepSeek V4.1, apenas $0,12.
El resultado más barato fue, de hecho, el que más me gustó.
@agentsky_dev permite comparar agentes, modelos y costes en el mismo dashboard antes de decidir.
We just launched Agentsky @agentsky_dev, world’s 1st Agent Market!
OpenRouter is for models. AgentSky is for agents.
Use 40+ agents—Claude Code, Codex, OpenCode, Hermes, Pi in your browser(even your phone!) or via one API. All without installing or setting up anything.
Hit Codex Astra’s weekly limit? Hand off to another agent such as OpenCode + DeepSeek V4.1 in browser without losing any context.
You can compare any agent + model directly in browser and that's how I found Astra costs $5.3 while deepseek v4.1 cost $0.12 on the same dashboard task. (I actually preferred deepseek)
Try it at agentsky.dev
If you can try any agent easily without install, which agent do you want to try next? Type the agent name and I will give $20 credit (you can use it for any model)
Arcane Matrix | AI retweeted
Misma tarea. Dos agentes. Una diferencia enorme en el coste.
Codex Astra: $5,30.
OpenCode + DeepSeek V4.1: solo $0,12.
Y lo más interesante: preferí el resultado más barato.
Eso es lo útil de @agentsky_dev: puedes comparar resultados y costes antes de elegir tu agente.
We just launched Agentsky @agentsky_dev, world’s 1st Agent Market!
OpenRouter is for models. AgentSky is for agents.
Use 40+ agents—Claude Code, Codex, OpenCode, Hermes, Pi in your browser(even your phone!) or via one API. All without installing or setting up anything.
Hit Codex Astra’s weekly limit? Hand off to another agent such as OpenCode + DeepSeek V4.1 in browser without losing any context.
You can compare any agent + model directly in browser and that's how I found Astra costs $5.3 while deepseek v4.1 cost $0.12 on the same dashboard task. (I actually preferred deepseek)
Try it at agentsky.dev
If you can try any agent easily without install, which agent do you want to try next? Type the agent name and I will give $20 credit (you can use it for any model)
I ran the exact same dashboard task through two different agents.
Codex Astra used $5.30.
OpenCode + DeepSeek V4.1 used only $0.12.
The cheaper one still delivered the result I preferred.
That’s the real value of @agentsky_dev: compare agents, models, and costs before committing.
We just launched Agentsky @agentsky_dev, world’s 1st Agent Market!
OpenRouter is for models. AgentSky is for agents.
Use 40+ agents—Claude Code, Codex, OpenCode, Hermes, Pi in your browser(even your phone!) or via one API. All without installing or setting up anything.
Hit Codex Astra’s weekly limit? Hand off to another agent such as OpenCode + DeepSeek V4.1 in browser without losing any context.
You can compare any agent + model directly in browser and that's how I found Astra costs $5.3 while deepseek v4.1 cost $0.12 on the same dashboard task. (I actually preferred deepseek)
Try it at agentsky.dev
If you can try any agent easily without install, which agent do you want to try next? Type the agent name and I will give $20 credit (you can use it for any model)
Arcane Matrix | AI retweeted
I ran the exact same dashboard task through two different agents.
Codex Astra used $5.30.
OpenCode + DeepSeek V4.1 used only $0.12.
The cheaper one still delivered the result I preferred.
That’s the real value of @agentsky_dev: compare agents, models, and costs before committing.
We just launched Agentsky @agentsky_dev, world’s 1st Agent Market!
OpenRouter is for models. AgentSky is for agents.
Use 40+ agents—Claude Code, Codex, OpenCode, Hermes, Pi in your browser(even your phone!) or via one API. All without installing or setting up anything.
Hit Codex Astra’s weekly limit? Hand off to another agent such as OpenCode + DeepSeek V4.1 in browser without losing any context.
You can compare any agent + model directly in browser and that's how I found Astra costs $5.3 while deepseek v4.1 cost $0.12 on the same dashboard task. (I actually preferred deepseek)
Try it at agentsky.dev
If you can try any agent easily without install, which agent do you want to try next? Type the agent name and I will give $20 credit (you can use it for any model)
Arcane Matrix | AI retweeted
Same task. Different agents. Completely different costs.
Codex Astra: $5.30.
OpenCode + DeepSeek V4.1: $0.12.
And the interesting part? I actually preferred the cheaper output.
@agentsky_dev makes these comparisons easy: same dashboard, same task, different agents.
We just launched Agentsky @agentsky_dev, world’s 1st Agent Market!
OpenRouter is for models. AgentSky is for agents.
Use 40+ agents—Claude Code, Codex, OpenCode, Hermes, Pi in your browser(even your phone!) or via one API. All without installing or setting up anything.
Hit Codex Astra’s weekly limit? Hand off to another agent such as OpenCode + DeepSeek V4.1 in browser without losing any context.
You can compare any agent + model directly in browser and that's how I found Astra costs $5.3 while deepseek v4.1 cost $0.12 on the same dashboard task. (I actually preferred deepseek)
Try it at agentsky.dev
If you can try any agent easily without install, which agent do you want to try next? Type the agent name and I will give $20 credit (you can use it for any model)
Arcane Matrix | AI retweeted
We just launched Agentsky @agentsky_dev, world’s 1st Agent Market!
OpenRouter is for models. AgentSky is for agents.
Use 40+ agents—Claude Code, Codex, OpenCode, Hermes, Pi in your browser(even your phone!) or via one API. All without installing or setting up anything.
Hit Codex Astra’s weekly limit? Hand off to another agent such as OpenCode + DeepSeek V4.1 in browser without losing any context.
You can compare any agent + model directly in browser and that's how I found Astra costs $5.3 while deepseek v4.1 cost $0.12 on the same dashboard task. (I actually preferred deepseek)
Try it at agentsky.dev
If you can try any agent easily without install, which agent do you want to try next? Type the agent name and I will give $20 credit (you can use it for any model)
Arcane Matrix | AI retweeted
Introducing AgentCloak: Use any AI without sharing your real data.
Chinese AI services, ChatGPT, Claude, doesn't matter.
You probably try to hide details before asking: different names, fake numbers, no address.
But then the answer's useless because the AI is missing actual context.
AgentCloak runs in your browser.
It swaps your sensitive info for realistic fakes before sending anything, then swaps your real info back into the response.
You get what you need.
The AI gets nothing about you.
Works entirely in-browser.
Already trusted by some of the biggest companies in the world.
Now free to use.
agentcloak.ai/