Rubén

Redditor claims Claude Code is nerfed for Pro/Max users vs Enterprise customers and the strategy is to use the paid plan users to generate hype on X and LinkedIn so companies would reach out to them.

hermes agent is on absolute fire. trending on today's news twice in the same day while eating into openclaw's lead every week. 5th most used AI agent in the world right now. the only one on that list that is fully open source from head to toe. open models, open harness, open memory system, open everything. no corporate chess pieces. no prompt mining. you own your cognition. nousresearch just shipped 7 pluggable memory providers in one release. the only open source lab releasing both the models AND the tools to run them locally. teknium merges contributions same day and ships like he doesn't sleep. the entire x timeline can confirm this. while other frameworks get bloated and slow down hermes agent gets leaner and faster. if you're still wondering why your model feels dumb on a bloated harness this is the answer. migrate, upgrade your cognitive tool, and stop being a pawn on someone else's chessboard.

Spain 🇪🇸 was just FORCED to eliminate VAT for self-employed people billing under €85,000 a year Every other EU country did this in January 2025. Germany, France, Italy... all of them. But of course, Spain said no I've been self-employed in Spain, so I know what quarterly VAT declarations feel like The modelo 303, the gestor fees, the Friday afternoons reconciling invoices... It's a real weight, especially when you're starting out and every hour counts. And that's not to mention that every Euro a client has to spend on VAT, he can't spend on you So when the EU passed the directive that would remove all of that for small freelancers, you'd think Spain would have jumped at it But... of course they didn't. For over a year, while 1.5 to 2 million Spanish autónomos kept filing and paying, their counterparts in every other EU country had already moved on The European Commission eventually took Spain to court over it. The potential fine was around €30 million And here's the part that's hard to swallow: by some estimates, Hacienda was collecting around €200 million a year from those small freelancers. Even with the fine, the math worked in their favor. So they kept saying no. They just don't care What finally changed it wasn't pressure from Brussels, but Junts threatening to vote against an unrelated energy crisis decree unless the government included the measure. Classic Anyway... It passed In practice: if you bill under €85,000 a year, you no longer have to charge VAT on your invoices or file quarterly declarations. The trade-off is you can't deduct input VAT on your expenses either, so it's more of an administrative relief than a straight tax cut. Whether it's worth it depends on your cost structure, but just the time you save alone makes this huge, imo For a lot of freelancers, especially those working with end consumers who can't deduct VAT anyway, this is genuinely good news But more importantly... this is too little, too late, as always. It took a court case and a political hostage situation to make it happen But hey, at least it's done That's Spain for you. The country is magnificent... but the system fights you every step of the way

@sudoingX @0xSero @sudoingX @0xSero better the sleep further you go, take care of your health, this is a long run let agents work late night. Run short research about not sleeping enough, it's scary :*

@0xSero 8 hours became a luxury. got a whole civilization to build

these are exciting times. the acceleration is insane. i slept at 2am and up by 6am. my mind wakes me up aching to get into the grind. it tells me treat a day like a year and a year like a decade. if i sleep 12 hours a day at 26, i won't make it. i need more hours in a day not less. building has never been this fun. feel the acceleration anon. good morning. i will conquer today like i did yesterday.

Hermes Agent now supports @plastic_lab's Honcho, @mem0ai, @openvikingai, @Vectorizeio's Hindsight, @retaindb, and @ByteroverDev memory systems! Try them now with `hermes update` then `hermes memory setup` We have rehauled our memory system to be much more maintainable and pluggable, so anyone can make their own memory system to build on top of Hermes easily and cleanly with a special class of plugin! Which memory system is your favorite?

Qwen3.5, MiniMax-M2.7 are incredible acts of kindness that I don't think will be with us from so much longer. Here's my update for you. > I have 20 GPUs at full utilisation right now. All these getting cooooompressed, no synthetic data All runs will be done in 9 days, if I don't get a catastrophic failure - REAP for: - GLM-5 - Qwen3-next-coder - Qwen3.5-122B - Qwen3.5-plus-397b - Browser-use - CUDA - Terminal-use - Coding - Math - Agentic trajectories - 30% my personal chat session history I am also removing refusals inspired by Prism. So no more I can't do this I can't do that blah blah Inference for local AI - Qwen3.5-262B-REAP - I've been using it exclusively in Parchi, perfect 100 tokens/s & 0 errors very good at browser use ----------------- Secret - Qwen3.5-27b - you will see when i'm done Targeting the following hardware levels: With full context 200-256k context in vllm, sglang, llama.cpp, exllamav3, and if people help MLX 16-32 GB - Qwen3.5-27b 32-48 GB - Qwen3-coder-next 48-128 GB - Qwen3.5-122B 128-256 GB - Qwen3.5-Plus-397B 196-512 GB - GLM-5.* I am training them on 22,000 samples at 16k context 352M of custom selected calibration datasets. My hope is to make the highest quality multimodal LLM compressions for this year. 20 GPUs running in parallel for the next 10 days - 8x H100s - Qwen - 4x B200s - GLM-5.* - 8x 3090s - Testing Once MiniMax-M2.7 is online 4 more GPUs will get to work.

@JEBistline x.com/NewPositivism/…. "Plants in 650 ppm CO2 flowered 14 days earlier and produced smaller pods with smaller and lighter seeds."

This is my favorite climate change chart. Japanese monks, aristocrats, and emperors kept meticulous records of cherry blossom festivals for 1,200 years and accidentally built the world's longest climate dataset.

@MilksandMatcha Vamos! XD

Giving away 5 Codex Pro plans Each person will get 3 months of free Codex Pro (highest tier). Winners will be selected from comments in 48 hours, comment below why you want it.

Casi me lo creo xD q bueno!

@bcherny What about the missing history bug since .89?

Today we're excited to announce NO_FLICKER mode for Claude Code in the terminal It uses an experimental new renderer that we're excited about. The renderer is early and has tradeoffs, but already we've found that most internal users prefer it over the old renderer. It also supports mouse events (yes, in a terminal). Try it: CLAUDE_CODE_NO_FLICKER=1 claude

There's also downsides: - Native cmd-f doesn't work. Instead, hit ctrl+o then / to search the transcript (or, use ctrl-r for reverse search) - Native copy-paste doesn't work. Instead, we copy to the clipboard by default when you make a selection. You can configure this in your settings.json to use ctrl+c instead, if you prefer. - Scrolling gravity varies by device, and we are in the process of tuning the physics to make it feel good on every terminal

@0xSero I think I can get used to this :)

Do you want to try Droid? I’m doing a giveaway 3 people will win 100M Factory credits each.Thats 5 months of their 20$ a month subscription. Winners selected randomly from comments in 48 hours.

I respect that the leaked source code for the @claudeai client is protected by copyright. But wouldn’t it be ok for me to train my LLM on it? You know, fair use and all that. Asking for a friend.

@DotCSV Madre mía, vaya hurto de feature chorra de hermes agent xD

Usad /buddy en Claude Code y mostradme vuestra mascota legendaria! Si a alguno le sale un pangolín que me lo pase con el cable link 👍

If you have a Thunderbolt or USB4 eGPU and a Mac, today is the day you've been waiting for! Apple finally approved our driver for both AMD and NVIDIA. It's so easy to install now a Qwen could do it, then it can run that Qwen...

@Rahatcodes if (is_negative) destroy_human

Claude Code has a regex that detects "wtf", "ffs", "piece of shit", "fuck you", "this sucks" etc. It doesn't change behavior...it just silently logs is_negative: true to analytics. Anthropic is tracking how often you rage at your AI Do with this information what you will

@jordimash 🙌 no offense, només compartia coses recents relacionades amb una conversa recent T'admiro infinit, disculpa si el to no ha estat adeqüat :-*

@enrub_en No sé si saps que soc el mantenidor de github.com/opennmt/ctrans… Ho dic perquè pots assumir que això ho sé

La realitat es amb els models Llama amb futur incert, Mistral i Olmo per darrere, Gemma 3 per casos petits, les grans opcions lliures on et calen models potents són principalment els models xinesos DeepSeek, Kimi o Qwen. ara.cat/internacional/…

@jordimash Ja sé que muntar un cluster de 8x3090 rtx no està a l'abast de tothom (jo el primer :p) però mica en mica ens apropem a executar coses GROSSES amb maquinari de consum x.com/0xSero/status/…

@enrub_en Qwen 3.5 té models petits (p ex 27B) fins a models grans de (+400B). Òbviament les prestacions no són les mateixes.