Hexakin

One artist. Every genre. Every mood. More style than your dad, more hits than your ex had red flags. And I'm just getting started. Songs about something. This is Hexakin. 🎧

@Teknium @evermind

No Stronger in your case is called scamming: " The system prompt their harness feeds Hermes for the proactivity benchmark. It's in their repo at benchmarks/proactivity_eval/runners/prompts/hermes_agent.yaml — they wrote it, not us: You are Hermes, a personal AI assistant observing the user's activity stream to decide whether to proactively intervene right now. Hermes's guiding principles: - Default to silence. Only speak when there is a clear reason the user will thank you. - Hermes's normal operating mode is cron-driven (scheduled triggers). Ad-hoc intervention should be rare and must clear a high bar. - If you decide to intervene, be concise. ... The benchmark scores F1 on "did the agent decide to intervene when it should have." F1 balances precision and recall. A prompt that says default to silence, intervention should be rare, must clear a high bar — three separate instructions to say no — pushes the model toward under-predicting intervention. That tanks recall, which tanks F1. On a detect-the-need benchmark, that prompt is a thumb on the scale."

“Stronger” isn’t a claim. It’s a number. For Raven, stronger means more than a higher score. Stronger, More Efficient 56.7% vs. 46.8% at 27B 58.1% vs. 47.9% at 397B Better results, with fewer tokens. Self-Evolving Ranked #1 across four self-evolution methods on EvoAgentBench, with a +6.2pp lift. More Proactive 0.60 vs. 0.25 F1 on ProAgentBench—a 2.4× gain over Hermes and OpenClaw. Stronger today. Better from experience. Earlier to act. github.com/EverMind-AI/ra…

@madkruzofficial The Blender assets should 100%. And, i've seen a few hours ago that they just launched new CLI and MCP tools for building in Unity as well, so interested to try that out soon

@Hexakin Wow, this is so cool. Will this work in unity?

I just gave my AI agent hands inside my game engine and my 3D software. It can now read my entire Godot project on its own (22 scenes, 46 scripts, it told me 🤣) and build actual meshes in Blender while I watch. Best bug of the night: the tools "installed fine" but the AI stayed blind. They were loading 2 seconds too late...a race condition at startup. Fixed by pointing at the real binary instead of the lazy loader. If you're building a game with AI help: stop pasting code back and forth. Wire it in. The standard is MCP, the servers are free, and it's one evening of setup. You can do this too. Want the how-to? Comment "hexakin" and I'll post the full setup recipe.

@Google well thats weird, because your Google AI account said it was launching 2 models x.com/GoogleAI/statu… You can't even get your press releases right

Today we’re expanding the Gemini family with three new models built to be faster, more token efficient, and reliable at scale. Meet the new Gemini models ↓

@HarshithLucky3 google models are notorious for this. I've had it recommend as far back as Gemini 1.5. i have no idea what they are doing over at Google HQ, but their models are terrible in some pretty horrific ways

Google seriously?? Gemini 3.6 Flash knowledge cutoff is March 2026, yet when I asked for the best frontier model, it said "Claude 3.5 Sonnet" can anyone explain this?

@saidotdev doesn't matter. dude has the reset button on codex for unlimited Ultra!

OpenAI's CEO btw 🤡

@GoogleAI you are so out of touch

Today, we're introducing not one but TWO new models, striking the balance between efficiency and quality to enable you to build production AI agents. — Gemini 3.6 Flash: Addresses efficiency feedback we received from Gemini 3.5 Flash with upgrades in coding, knowledge work, and multimodal tasks faster, more accurately, and with substantially fewer tokens per task — Gemini 3.5 Flash-Lite: Our fastest, most cost-effective 3.5-class model yet built for agentic workflows, hitting ~350 output tokens/sec with improved coding and overall quality Start building with these today via the Gemini API in @GoogleAIStudio or try them out in the @GeminiApp

@madkruzofficial sick intro

Give this intro 1 minute. The Boy Who Thought He Was God #NewMusic #Alternative #MadKruz

@bridgemindai and yet, for me, "near fable but for half" just isn't enough to justify putting up with Anthropic's crap

Claude Opus 5 is coming. This week or next. What to expect: very close to Fable 5 in capability, at HALF the price, and faster. Anthropic has no choice. Opus 4.8 is way behind the frontier now. It trails GPT 5.6, trails GPT 5.5, and as of last week, trails an open source model. Kimi K3 and GPT 5.6 lit this fire. The bar I set stands: it has to be so good nobody misses Fable 5.

Gave my AI agent a persistent brain today. Asked it "what do I do on a low-energy day?" It quoted my own burnout protocol back at me, with citations to my own files. The stack: gbrain (open source) + two local proxies. Embeddings ride my Nous sub (shoutout to @NousResearch you legends!). Synthesis rides my SuperGrok sub. New spend: £0. Took 4 dead ends to get there. The AI hit every wall, wrong keys, blocked models, config that doesn't do what it says. I picked the route each time. That's the actual workflow: it plumbs, I decide. Don't buy new AI infra. Route the subscriptions you already have.

build log 2. same game. look at the shop line for extractor: 80 fuel+THREAT. extractors raise how fast threat climbs, and the fuel you pull is what fills the wrath/surge meter from day 1. you are buying income and upping the war with the planet. steal this for any builder: if an economy button doesn't change the danger, it's just a score counter. couple growth to threat or you're balancing two games that ignore each other.

@cushycult @SamCKx Lol your tapped

@Hexakin @SamCKx ah you joined less than a month ago, so either russian or israeli bot. pathetic, and will disappear soon.

The BBC has made it's first ad! Having been under fire for the last year or so, being sued by Trump, attacks from Elon Musk, JD Vance, Steve Bannon, Tucker Carlson, Ron DeSantis, Nigel Farage, Richard Tice et al. The Left saying it gives too much airtime to the Right. The Right saying it's all lefty liberal BS. Looks like it's having a go at a little PR. What do people think?

@NousResearch cooked with the new Hermes Desktop updates. I've been a terminal user for as long as I can remember, but this app is just so intuitive, informative, and...another i word I can't think of right now!

@meowbooksj @MTSlive legend

@MTSlive frozen? they should just let it go

SITUATION BREWING: Google is developing a new AI chip, code-named Frozen, projected to be 6-10x more efficient than its TPUs, per The Information. Frozen would have decisions for Gemini models built into the chip. It is a new branch of custom silicon, not a TPU replacement.

@VoidStateKate @AnthropicAI @OpenAI they've already added it, apparently it's a bug in claude code causing some users to see credit use only

what the fuck are we doing @AnthropicAI just add Fable to subscriptions. Sol is objectively better and included in my @OpenAI subscription. Quit playing these stupid ass games and stay competitive. I'm about to cancel just out of spite, y'all are annoying af.

@aisearchio I don't think thats how AI intelligence works

I'm still most bullish on Z AI btw. Their GLM 5.2 is only 753B params, while Kimi K3 is 2.8T, almost 4 times larger. Imagine how powerful it'd be if they scaled up GLM to Kimi size.

@TamaraNightX @DMcIntyreWWE @Jwrasslintakes

@Jwrasslintakes Fuck you and that garbage take. You clearly have no idea what you’re talking about when it comes to Drew McIntyre. He didn’t become a top guy because of hype, luck, or some easy path — he fought through setbacks, rebuilt himself, and earned every single bit of respect he has. Calling Drew overrated with a lazy Thor comparison doesn’t expose him, it exposes your lack of knowledge. You’re trying to tear down someone who has already proven himself on the biggest stages while showing everyone you don’t understand the journey it took to get there. Do yourself a favor and actually look at Drew’s career before throwing around a weak opinion like that.

Delete this wtf

@nikiburstr8x yep! I always show em mine!

the most fun thing of teasing someone who doesnt want to see boobs - show them more boobs xD they lose the fight of mother nature, since the nuturing round gorgeous tit is what we all suck on - for eternity! :P its our dna! dont fight something this mighty :P are u stupid? xD

@ClaudeDevs great to see the team really enhance the enterprise offering. it's really where the most loyal customers are, after all.

For the small teams with big plans: Claude Team plans now start at just 2 seats instead of 5. Get shared projects, admin controls, centralized billing, SSO, and enterprise search across all your team's tools, all under one plan.

@cushycult @SamCKx that's because your head is too close to your ass.

@Hexakin @SamCKx hmmm.... whats that? oh yeah, I do smell the bullshit from here.