My dev workflow for now does not use any local models and uses only the harnesses provided by the big labs. So Codex for OpenAI models and Claude-code for Anthropic models. Are harnesses like pi that can be model agnostic worth a try?

Okay I have been recommended by a few to try @pear_protocol to play $XMRZEC cross perps Never used them before so will check as nowhere else offers it Anyone else used them? Would like your opinion.

Opus 5.5 is next level. I’m genuinely impressed by the three.js details. I generated a playable boat scene through Japanese landscapes with dynamic weather, day and night lighting, realistic textures and 3D characters. Live site: valley.mengto.here.now The water reflections, physics, scenery and architecture are insane. I could look at this all day.

We have a beautiful gift for the Pear Army. This Sunday, on the second anniversary of our token. Keep your eyes peeled. 🍐

Yesterday marked one month anniversary of @DarkbloomAI moving to paid tier on OpenRouter. We started with 4B tokens a day and today we are excited to be serving 20B+ tokens a day across 8 models. We have grown our provider base to 1.1K+ active at all times and plan to expand it much higher as we expand the model base and more powerful M5 chips are getting shipped over next few weeks. You can now run Darkbloom on any Macs (including corp) without our MDM. Last week, we piloted our new higher grade privacy model which allows us to bind machines to stronger trust guarantees leading to Apple's certificates. You must be on MacOS 27 to be part of this and you will be soon required to be on OS 27+. During last 2 weeks, we have shipped a bunch of features to increase the QoL for our providers which includes Global Payouts, better routing, MLX(.)Fast inference engine improvements etc. Thank you everyone for being early to our network and supporting us through all the journey. I also want to thank the OpenRouter team (@pingToven, @shashankgoyal95, Eddie, Will, @OpenRouterMatt) for making this happen and continuing to support us through adding new models. This week we're at our company offsite + retreat planning through all the things we need to ship in the coming weeks and months. We will share a much detailed update after this. If you are on Darkbloom, make sure you're part of the Slack community.

the ETH/BTC ratio is a good indicator for how cheap ethereum:native still is. anything below ~0.165 is cheap anything above ~0.165 is fair value 😘

confession: I see privacy as more of a "eat your vegetables" type of thing in crypto it's obviously needed for crypto to scale to the next level, especially for corporate use cases, but it's never intrinsically excited me, it's more of a chore maybe because I'm an 'open book' kind of person myself and don't really do dirt, IDK. . . what I find exciting is the programmability, trust minimization, ability to bind people to hard obligations without a legal system, reduction of transaction costs, automation, replacing obligations with incentives, etc. privacy is obviously an enabler of all that (in a hostile world you can't be free if all your enemies can see everything you do and use it against you) but IDK, I just never found it intrinsically exciting

These system prompts are getting crazy.

There are only two truly decentralized L1s with zero counterparty risks. ETH and BTC. With BTC doing nothing more than moving BTC. And ETH bringing every imaginable usecase onchain. 25k 100k2030 1M longterm. It's programmed.

Together, the MLX(.)fast community has made Qwen 3.8 Flash nearly 2x faster on Apple Silicon! We're ready to bring it to @DarkbloomAI: an open network of local Mac machines providing inference to the world. One thing remains: the community flagged that its license requires a separate agreement for commercial model serving, so we're holding the launch until that's in place. We believe this is a great opportunity for the local community: one where we make Qwen models faster and more accessible, and the people running them share in the value they create. .@Alibaba_Qwen @QwenDevs, we'd love to work together on this. If anyone else knows someone we can talk to, we'd love to have that conversation. Let's make Qwen 3.8 Flash on Darkbloom a reality!

Today, 0x published an interesting and important article about an emerging phenomenon, first seen in PropAMMs, now in UniV4 hooks. To summarize: malicious makers alternate between very tight and very wide quotes. Tight quotes get the venue selected by the aggregator, but then when the user's trade gets routed to the venue the quote is wide. Slippage settings (i.e. limit price) protects some users, causing the trade to just fail, but many users set their slippage settings far too permissively. Those users get taxed brutally - the article highlights some hooks alternating between 0 and 18% fees. (A few months ago, Solana posted an really excellent article about this same phenomenon occurring on Solana aggregators, although with upper-end fees more like 1% instead of 18%. They quietly took the article down a few days later.) The promise of open systems is elimination of middlemen and hidden taxes. Aggregator spoofing represents a huge hidden tax, and it can only truly be solved by fully-on-chain order routing. That in turn requires plentiful computation and flexible account access (i.e. no access lists, no pre-specification of touched accounts like in Solana). You basically need a really performant and efficient EVM. On top of that, you need a well-designed on-chain routing protocol. Who's building this?

Darkbloom is back in action; Yesterday, we went from free tier to paid tier on OpenRouter. We have fulfilled on track close at 4.5B tokens served and $102K ARR. We are open network of Mac machines (250 online right now). On average, people are earning $120-200 per month per machine. If you have a Mac and it's idle most times: put it on Darkbloom(.)dev; Tip: run the Gemma 4 26B model.