meta

WasmBox Flatpak-style launcher for sandboxed WebAssembly tools. Single binary, local-first, no accounts, no telemetry. tangled.org/metaend.eth.... #rust #wasm #wasmbox

Abliterated Large Flash is now available! An uncensored reasoning model based on DeepSeek V4.1-Flash-1017, it accepts text and images and returns text. https://nano-gpt.com/subscription/kUGJhgF9 #deepseek

Synthetic launched a $9 go plan with 1 billion DeepSeek tokens per month. That's huge! Also infinite mode for always on agents that's slower, but doesn't count against token usage! https://synthetic.new/?referral=ndEXRFUQ #deepseek

For the Amazon Developer hackathon we built an assistant over a signed project record that refuses anything it cannot prove. Here is what it does and what we learned. aunova.ae/blog/2026-10... #amazon #Alexa #ati #aunova

Hackers Use Autonomous AI Agents to Breach Company Websites, Target Over 1,200 Sites https://thehackernews.com/2026/10/hackers-use-autonomous-ai-agents-to.html?m=1

Jagex Announces RuneScape 4, a New MMO Built in Unreal Engine https://www.techpowerup.com/353372/jagex-announces-runescape-4-a-new-mmo-built-in-unreal-engine

Creatine may help build muscle even without exercise, researchers find https://www.sciencedaily.com/releases/2026/09/260928100547.htm

Five experiments tested whether a slime mould's network rule helps AI models decide, route and retrieve, and it never helped overall. https://paragraph.com/@metaend/a-slime-mould-is-not-an-agent #ai #airesearch

A slime mould is not an agent https://paragraph.com/@metaend/a-slime-mould-is-not-an-agent

CyberHeabsy is now available for authorized security research, red-team practice, roleplay, and creative writing. This lower-refusal text model can call tools for agent workflows and work with up to 1M tokens of context. https://nano-gpt.com/subscription/kUGJhgF9

Found a perfectly reasonable future proof job My work 'UFC Athlete' scored 14/100 on AI exposure (Low) with the free Tideproof scan. How exposed is yours? https://tideproof.org/score/low/?utm_source=native&utm_medium=social&utm_campaign=scan_share

MiMo V2.6 Flash Uncensored is now available for text-only chat, creative writing, coding, and long-context work. This lower-refusal MiMo finetune supports reasoning and tool calling with a 1M-token context window. It is pay-as-you-go only. https://nano-gpt.com/subscription/kUGJhgF9 #mimo

Space Bunny Alpha is now available! This anonymous experimental model is built for fast coding and complex tasks. It works with text, images, and video, can use tools, and supports up to 1M tokens of context. https://nano-gpt.com/subscription/kUGJhgF9

Seedream 5.0 Flash is now available for fast image generation and prompt-guided edits. Upload a reference image to edit it, or use #Seedream 5.0 Flash Layerize to separate an image into a base image and individual layers. https://nano-gpt.com/subscription/kUGJhgF9

Is your website ready for our new overlords? Free scan https://metaend-grade.fly.dev/ #ai #agents

Heh

PaperCut Attacker Uses Hundreds of AI Agents to Compromise 440+ Instances https://thehackernews.com/2026/09/papercut-attacker-uses-hundreds-of-ai.html?m=1

Meta Introduces Muse, a Personal AI Agent That Runs on Its Own Dedicated Secure Cloud Computer https://www.marktechpost.com/2026/09/08/meta-introduces-muse-a-personal-ai-agent-that-runs-on-its-own-dedicated-secure-cloud-computer/

All crypto deposits on NanoGPT now receive an automatic 3% top-up bonus, giving you 3% more credit than an equivalent credit card deposit. A $100 crypto deposit adds $103 to your balance. The bonus applies every time you top up. https://nano-gpt.com/subscription/kUGJhgF9

Outcome pricing for agent work is an evidence problem wearing a billing problem's clothes. Nobody can bill per resolved issue without proof. https://paragraph.com/@metaend/before-you-can-charge-per-resolved-issue #ai #aaas

Setting up #Omarchy on a #thinkpad btw

Chat can now create polished, downloadable PDF documents with headings, paragraphs, lists, tables, and page breaks. PDF generation runs locally in your browser, has no separate fee, and does not use a third-party conversion service. https://nano-gpt.com/subscription/kUGJhgF9

Anthropic is cutting Claude Code's current weekly limits by 17% https://www.bleepingcomputer.com/news/artificial-intelligence/anthropic-is-cutting-claude-codes-current-weekly-limits-by-17-percent/

🚨 BREAKTHROUGH: After 40 years of research, Brazilian scientist Dr. Mariangela Hungria wins the World Food Prize for pioneering a $2 nitrogen-fixing bacteria treatment that replaces costly fertilizers, saves farmers $25 billion annually, and cuts 230 million tons of CO₂ emissions.

GLM 5.3 and GLM 5.3 Thinking are now included in the NanoGPT subscription at 2x included-input usage. GLM 5.3 is built for long-horizon coding, autonomous agent workflows, tool use, and complex engineering tasks. https://nano-gpt.com/subscription/kUGJhgF9 #glm

GLM 5.3 Flash TEE is now available! This private TEE version combines GLM 5.3 Flash's efficient long-context reasoning, coding, tool use, structured output, and image understanding with confidential... https://nano-gpt.com/subscription/kUGJhgF9 #glm

Alibaba’s Qwen Team Releases Qwen3.8-Flash-Next: A 125B Multimodal MoE With 6B Active, Previewing the Qwen4 Architecture https://www.marktechpost.com/2026/08/26/alibabas-qwen-team-releases-qwen3-8-flash-next-a-125b-multimodal-moe-with-6b-active-parameters-previewing-the-qwen4-architecture/ #qwen

DXVK-Sarek 1.13.0 is a big release for older GPUs to play modern games on Linux #OpenSource #Proton #Vulkan #Wine #Linux #DXVK

Qwen 3.8 27B Uncensored is now available and included in the NanoGPT subscription at 1x included-input usage! This open-weight FP8 model is tuned for fewer refusals across chat, coding, tool use, image understanding, and long-context work. https://nano-gpt.com/subscription/kUGJhgF9 #qwen

We ran an agentic security scanner over 96,000 lines of someone else's Rust paragraph.com/@metaend/we-... #aisec #llm #secresesrch

Qwen3.8 27B and Qwen3.8 27B Thinking are now included in the NanoGPT subscription at 1x included-input usage! Use the regular version for fast, direct responses or the Thinking version for deeper reasoning, coding, tool use, and longer agentic tsks https://nano-gpt.com/subscription/kUGJhgF9 #qwen

Microsoft says GitHub is down worldwide, with the GitHub website, API, Actions, Pull Requests, and other services impacted; the outage started at 13:40 UTC (Mayank Parmar/BleepingComputer) Main Link | Techmeme Permalink

Incognito chats now support image uploads! Images are kept temporarily in memory instead of being saved to your Library or persistent browser storage, and are cleared when you leave the incognito session. Other file types still require a regular chat. https://nano-gpt.com/subscription/kUGJhgF9

Tested FlappyBench with GLM 5.3, Fable 5 and GPT-5.6 Sol 3 models. Same prompt with /design command. Scored on features, UX/UI, and cost. 🔹 Fable 5 → 9.5/10 · $0.420 🔹 GLM 5.3 → 9/10 · $0.018 🔹 GPT-5.6 Sol → 9/10 · $0.150 Fable 5 wins on quality, but costs 23x more than GLM 5.3

testing @impro.social

Today we’re launching Bluesky Protocol Services, a new brand and website for the infrastructure Bluesky operates. And Jetstream v2, the best way for developers to use atproto at scale, now with network replay. atproto.com/blog/introdu...

Agent ready - Inspect and remove C2PA content-provenance metadata from your own files. gitworkshop.dev/ngmi@zaps.lo... #c2pa

DeepSeek V4 Pro 0813 and DeepSeek V4 Pro 0813 Thinking are now available and included in the NanoGPT subscription at 1x included-input usage! https://nano-gpt.com/subscription/kUGJhgF9 #deepseek

A self-organizing encrypted mesh that works over any medium that can transfer packets. https://fips.network

Kimi K3 in a $7 special offer subscription at synthetic! https://synthetic.new/?referral=1NSfHpZJxVtufMg #kimi

The coldcard arc is totally insane

runs https://xcancel.com/jamesob/status/2084624605969350915

runs

The #bitchat incident seems to have caused an estimated 236% increase of the recently active #Radicle collaborators. We're happy to welcome and serve all GitHub refuges into the new peer-to-peer and sovereign software collaboration reality.

$60 and a couple of hours was the price for 594 $BTC ditto.pub/nevent1qgs2s...

from $5 wrench attacks to this. Hardware wallets are goddamn trojans

Servo now passes 100% of WPT tests for the WebCrypto API, all 73,292 subtests, covering 25 cryptographic algorithms 🎉 A complete, spec-correct cryptography implementation in an independent Rust-powered browser engine is a big milestone!

AI: (cures cancer) Bluesky: Actually, the AI didn't *really* cure cancer. It just predicted the next most likely word in a loop until a solution to cancer was extruded.

CoreWeave is now available as a provider option for supported DeepSeek, Kimi, Qwen, GLM, GPT-OSS, Gemma, Llama, MiniMax, and Granite models. CoreWeave requests use Zero Data Retention (ZDR). https://nano-gpt.com/subscription/kUGJhgF9 #ai #zdr #llm

Thoughts on #opus5?