deckbsd🌱🌐 @deckbsd

Thanks to @nostarch for sending me an advanced copy of this, evenings reading sorted 👍

‼️🚨 He's back: Nightmare Eclipse just dropped RoguePlanet, a new Windows Defender local privilege escalation 0day PoC. The RCE paths broke after Microsoft's Defender patch. NE suspects the BitLocker bypass may still work but isn't certain. He has a new GitHub btw, let's see how long the account will last: github.com/MSNightmare/Ro…

keyboard_arrow_left Previous keyboard_arrow_right Next

MTP is up, test it out huggingface.co/Mia-AiLab/Qwab…

Mia @MiaAI_lab

4 days ago

I've published Qwable 3.6 27b to HF Qwable = Qwen + Fable. Based on Qwen 27B with Fable-style reasoning and assistant traces. Grab it now and test it yourself. huggingface.co/Mia-AiLab/Qwab…

71 116 1K 157K 1K

my 8 GB VRAM gaming laptop is absolutely going to hate me for this. but I still did it. ran a 31b dense model (Gemma 4 31b Q4) with only 8 GB VRAM last week I ran Gemma 4 26B A4B a mixture of experts model on my RTX 4060 and hit 25–28 tokens/sec using llama.cpp's new MTP support. smooth. snappy. but MoE has a secret: it only activates 4B parameters per token despite having 26B total. that's why it flies. so the real question started haunting me. what if I throw a full, no tricks, every parameter fires on every token, 31B DENSE model at the same machine? # Hardware: GPU: NVIDIA RTX 4060, 8 GB VRAM RAM: 16 GB CPU: Intel Core i7 H Laptop. Gaming. Modest. The model: gemma-4-31B-it-qat-UD-Q4_K_XL.gguf (model's unsloth huggingface link in the comments) This is Google DeepMind's flagship dense model in the Gemma 4 family that can run on single consumer GPU. It packs a hybrid attention architecture, supports up to 256K context natively, and is QAT (Quantization Aware Training) optimized, meaning it retains far more quality than standard post training quants at the same bit depth. This is NOT the MoE. This is 31 BILLION dense parameters, every single one of them loaded. # the flags I used: -m gemma-4-31B-it-qat-UD-Q4_K_XL.gguf -cnv --spec-type draft-mtp --spec-draft-model mtp-gemma-4-31B-it.gguf --spec-draft-n-max 8 --spec-draft-p-min 0.6 -c 6000 -v Multi Token Prediction (MTP) is still active here. Separate draft GGUF required, same as the 26B setup. # Results: → Decode: ~3 tokens/sec → Prefill: ~2 tokens/sec → Context: 6000 tokens → Hardware crying quietly in the corner: yes so is 3 tps actually usable? For real time back and forth chat? Not ideal. You're not having a fluid conversation at 3 tps. but slow ≠ useless. And this is where it gets genuinely interesting. think about how senior devs actually work in a real team. But when something is architectural, deeply complex, or needs serious reasoning? they walk down the hall and escalate to the senior. That's exactly the local AI agent architecture this unlocks: → Fast orchestrator model (Gemma 4 26B MoE at 25+ tps) handles routing, simple queries, tool calls, memory. The junior dev. → Gemma 4 31B dense is the senior, called only when the fast model genuinely hits a wall. Hard multi step reasoning. Complex code generation. Deep architectural decisions. The agentic loop stays fast. Only the hard hops touch the 31B. That's a legitimate production grade local AI architecture on a budget hardware. (requires 2 8gb gpus) other workflows where 3 tps is completely fine: - overnight batch jobs. summarize documents, extract structured data, review code. Fire it off. Sleep. wake up to results. - One shot deep reasoning - Silent code audit loops, you write and test, the 31B reviews diffs and flags issues in the background between your sprints - Any workflow where output quality > output speed A few weeks ago, nobody was running a 30B+ dense model on a single consumer GPU with 8 GB VRAM. At all. Now we're doing it on an Intel i7-H gaming laptop with a NVIDIA RTX 4060, thanks to llama.cpp + QAT quants + MTP speculative drafting. Google DeepMind said the Gemma 4 31B targets "consumer GPUs and workstations." They were not exaggerating. The hardware bar to run serious frontier class models locally keeps dropping. the tools are here. the models are here. you just have to be willing to abuse your laptop a little. what workflows would you actually run on a local 3 tps 31B dense model? genuinely curious. drop it below.

Alok @analogalok

2 weeks ago

Run Gemma 4 26b MTP on 8 GB VRAM GPUs at 25+ tokens/second. Flags included! local llm space is moving at terminal velocity. only 3 days ago google released gemma 4 26b a4b qat quants. more efficient than before, ran on 8gb vram at 20 tok/sec. and now just a few hours ago,

31 47 337 197K 519

My first optical satellite observation 48865 COSMOS 2550 with STVID, this will be an addition to my Doppler measurements. github.com/cbassa/stvid/

OptimizerDuck, open-source tool yang bikin lo uninstall CCleaner. Ini bukan cuma "cleaner" biasa. Dia gabungin 30+ tweak Windows dalam satu app, dari matiin telemetry, block bloatware, sampai GPU registry tweaks yang biasanya lo harus edit manual. Yang gue suka, dia kasih risk rating buat setiap tweak. Jadi lo tau sebelum apply, bukan asal pencet terus nyesel. Fitur yang kerasa banget: - Disable Windows telemetry, Cortana, Copilot, advertising ID - Startup manager, lo bisa matiin semua app yang auto jalan pas boot - Service host tuning based on RAM lo - Custom high-performance power plan - Keyboard latency reduction buat gaming Dan semua reversible. Ga cocok? Balikin lagi. Yang paling penting adalah, portable .exe. Download, langsung jalan. Ga install, ga registry sampah. Bisa di USB stick. Dan dia open-source, bukan freemium yang nanti nagih. 2.3k stars, aktif dikembangin (commit terakhir 2 hari lalu). Support 8 bahasa termasuk Indonesia? Engga. Tapi EN, Vietnam, China, Korea, Prancis, Spanyol, Rusia ada. Link: github.com/itsfatduck/opt… CCleaner Pro $40/tahun buat apa kalau ini ada.

Easily fuzz log files using a wordlist of dates 🗓️  Here’s how: 1️⃣ Use ffuf to find an active log directory 2️⃣ Grab a wordlist of dates 3️⃣ Run ffuf -w dates.txt -u /log/FUZZ.log -ac -mc 200 Want to try this technique out? Hack our Content Discovery Hub now! Link in comment 🔗

🖥️ Best Local LLMs for Consumer GPUs — llama.cpp Guide (June 2026) What I actually run on consumer hardware right now. Every model below runs via llama.cpp with a simple one-liner — no Docker, no Python env, no cloud. ━━━ 8-16GB VRAM ━━━ 🔹 Gemma 4-12B (Google) • Smartest model in this size class — competes with stuff 2× bigger • Unsloth's MTP GGUFs: 162 tok/s vs 52 tok/s normal (3× speedup) • Minimum 8GB VRAM recommended for Q4_K_M quant • GGUF → huggingface.co/unsloth/gemma-… 🔹 LFM2.5-8B-A1B (LiquidAI) • Hybrid MoE, only 1B active params — absurdly fast for its size • Perfect for 8-12GB cards, MacBooks, or anyone on a tight budget • GGUF → huggingface.co/LiquidAI/LFM2.… ━━━ 16-32GB VRAM ━━━ 🔹 Qwen3.6-27B (Qwen) • Scored 1.00 on tool-efficiency benchmarks — best local agent available • 40 deterministic tasks, 32k/128k context needle tests — all passed • GGUF → huggingface.co/unsloth/Qwen3.… • MTP version (faster) → huggingface.co/unsloth/Qwen3.… 🔹 Qwopus3.6-27B-v2 (Jackrong) • Best quantization of Qwen3.6-27B — topped 5 agent & coding benchmarks (1200 samples) • If you're running Q4, this is the one to grab • GGUF → huggingface.co/Jackrong/Qwopu… • MTP version → huggingface.co/Jackrong/Qwopu… 🔹 Gemma 4-31B QAT (Google/Unsloth) • QAT variant with MTP draft head: 76-125 tok/s (1.67× speedup) • Excellent for multi-agent / subagent workflows • GGUF → huggingface.co/unsloth/gemma-… 🔹 Nex-N2-Mini (Nex AGI) • Post-train of Qwen3.5-35B-A3B — MoE with only 3B active params • Fits on 16GB+ VRAM, overflow loads from system RAM • Adaptive thinking saves ~20% tokens with no quality loss • For deep multi-step reasoning, nothing in this size comes close • GGUF → huggingface.co/sjakek/Nex-N2-… ━━━ Quick Picks ━━━ • 16GB all-rounder → Gemma 4-12B with MTP GGUFs • 32GB all-rounder → Qwen3.6-27B / Qwopus-v2 • Agents & tool use → Qwen3.6-27B or Qwopus Q4 • Deep reasoning → Nex-N2-Mini (MoE, fits 16GB+) • Tight budget → LFM2.5-8B-A1B • Cheapest full build: 1× used RTX 3090 (24GB) + rest of PC ≈ $1000-1500 ━━━ Setup on Windows ━━━ 1. Download llama.cpp → github.com/ggml-org/llama… (latest .zip) 2. Extract to any folder (e.g. C:\llama.cpp) 3. Download a .gguf from the links above (Q4_K_M or Q5_K_M for best quality/speed balance) 4. Run one of the commands below depending on your hardware ━━━ Launch Commands ━━━ SINGLE GPU — Standard model (no MTP): llama-server.exe ^ -m C:\models\Qwen3.6-27B-Q5_K_M.gguf ^ --ctx-size 180000 ^ --flash-attn on ^ --cache-type-k q4_0 ^ --cache-type-v q4_0 ^ --batch-size 1024 --ubatch-size 512 ^ -ngl 100 ^ -np 1 ^ --port 8080 ^ --jinja SINGLE GPU — MTP model (faster inference): llama-server.exe ^ -m C:\models\Qwen3.6-27B-MTP-Q5_K_M.gguf ^ --ctx-size 180000 ^ --flash-attn on ^ --cache-type-k q4_0 ^ --cache-type-v q4_0 ^ --batch-size 1024 --ubatch-size 512 ^ --spec-type draft-mtp ^ --spec-draft-n-max 3 ^ -ngl 100 ^ -np 1 ^ --port 8080 ^ --jinja DUAL GPU — Split across two cards: llama-server.exe ^ -m C:\models\Qwen3.6-27B-Q5_K_M.gguf ^ --ctx-size 180000 ^ --flash-attn on ^ --cache-type-k q4_0 ^ --cache-type-v q4_0 ^ --batch-size 1024 --ubatch-size 512 ^ -ngl 100 ^ --tensor-split 0.55,0.45 ^ --main-gpu 0 ^ -np 1 ^ --port 8080 ^ --jinja DUAL GPU + MTP + Vision (multimodal): llama-server.exe ^ -m C:\models\Qwen3.6-27B-MTP-Q5_K_M.gguf ^ --ctx-size 180000 ^ --flash-attn on ^ --cache-type-k q4_0 ^ --cache-type-v q4_0 ^ --batch-size 1024 --ubatch-size 512 ^ --spec-type draft-mtp ^ --spec-draft-n-max 3 ^ -ngl 100 ^ --tensor-split 0.60,0.40 ^ --main-gpu 0 ^ -np 1 ^ --port 8080 ^ --jinja ^ --mmproj C:\models\mmproj-F16.gguf ━━━ Parameter Breakdown ━━━ -m Path to your .gguf model file. Change this to wherever you downloaded it. --ctx-size 180000 Context window in tokens. 180k = huge context for long conversations or big codebases. Reduce to 32768 or 65536 if you don't need long context — uses less VRAM. --flash-attn on Flash Attention — dramatically speeds up inference and reduces VRAM usage. Works on RTX 30xx/40xx/50xx. Always enable this. --cache-type-k q4_0 / --cache-type-v q4_0 Quantizes the KV cache (key/value attention cache) to 4-bit. This is what makes 180k context fit in VRAM. Without it, huge contexts eat all your memory. Quality impact is minimal — this is a free performance win. --batch-size 1024 / --ubatch-size 512 batch-size = how many tokens are processed in one forward pass (throughput). ubatch-size = micro-batch actually sent to the GPU per step. Higher = faster prompt processing but needs more VRAM. If you run out of VRAM, lower these (e.g. 512/256). -ngl 100 Number of layers to offload to GPU. 100 = all layers on GPU (full offload). This is what you want if the model fits in your VRAM. If it doesn't fit, reduce this (e.g. -ngl 40) — remaining layers run on CPU/RAM. --tensor-split 0.55,0.45 How to split model layers across multiple GPUs. Values are ratios. 0.55,0.45 = GPU 0 gets 55% of layers, GPU 1 gets 45%. Adjust based on your VRAM — give more to the card with more memory. Example: 0.70,0.30 for a 24GB + 12GB setup. Not needed for single GPU setups. --main-gpu 0 Which GPU handles the batch computation (the "orchestrator"). Set to 0 (your primary GPU). The other GPU(s) handle their assigned layers. Minor performance impact — usually just leave it at 0. -np 1 Number of parallel slots (concurrent requests). 1 = one user at a time. Increase to 2-4 if you want multiple clients connected simultaneously. Each extra slot uses additional VRAM for its own KV cache. --port 8080 Which port the server listens on. Change if port 8080 is busy. --jinja Enables Jinja2 template processing — required for proper chat formatting. Most modern models expect this. Always include it. --spec-type draft-mtp Enables Multi-Token Prediction (MTP) speculative decoding. Only works with MTP GGUF models (downloaded separately). The model predicts multiple tokens at once and verifies them — big speed boost. --spec-draft-n-max 3 How many tokens the MTP draft head proposes per step. 3 is a good default. Higher = potentially faster but more VRAM and may reduce quality. --mmproj Path to the multimodal projector file (for vision models). Enables image understanding — paste screenshots into the web chat. Only needed if you want vision capabilities. Omit for text-only use. ━━━ Your Hardware → Your Command ━━━ Single GPU (8-24GB VRAM): Use the "Single GPU" command. Change -m to your model path. 8GB card → Gemma 4-12B Q4 or LFM2.5-8B 12GB card → Gemma 4-12B Q5/Q6 16GB card → Gemma 4-31B QAT Q4 or Nex-N2-Mini 24GB card → Qwen3.6-27B Q4/Q5, Qwopus-v2, Gemma 4-31B QAT Q5/Q6 Dual GPU: Use the "Dual GPU" command. Adjust --tensor-split based on your VRAM ratio. 24GB + 24GB → --tensor-split 0.50,0.50 24GB + 12GB → --tensor-split 0.70,0.30 24GB + 8GB → --tensor-split 0.75,0.25 Want speed? Use MTP versions of models with the "MTP" commands. Want vision? Add --mmproj with the projector file from the model's HuggingFace repo. 5. Once running, you get: • Web chat UI → http://localhost:8080 • OpenAI-compatible API → http://localhost:8080/v1 • Playground → http://localhost:8080/playground ━━━ Why /v1 API Is the Killer Feature ━━━ One local endpoint replaces your entire cloud API bill. The /v1 endpoint is drop-in OpenAI-spec compatible — every tool that speaks OpenAI just works. No custom code, no glue layer. Works out of the box with: • IDEs: Cursor, Continue, Windsurf, Cline, Roo Code • CLI tools: aider, Open Interpreter, OpenCode • Frameworks: LangChain, LlamaIndex, LiteLLM • Any OpenAI SDK (Python, Node, Go, Rust) Why this beats cloud APIs: • 100% private — code never leaves your machine • $0 per token — no rate limits, no quotas, no surprise bills • Works fully offline • Zero telemetry, no training on your data • Swap models by dropping in a different .gguf — no app changes needed • Run 32k–128k context windows without burning money Good combos: • Cursor + Qwopus-v2 → near-frontier quality, zero API cost • Continue + Qwen3.6-27B → best local coding agent • aider + Gemma 4-12B MTP → 162 tok/s, feels instant • OpenCode + Nex-N2-Mini → deep reasoning on 16GB Set any OpenAI-compatible client to your local endpoint: set OPENAI_API_KEY=sk-dummy (any non-empty string works) set OPENAI_BASE_URL=http://localhost:8080/v1 # every OpenAI-compatible tool now hits your local GPU Shoutouts: @0xSero @rS_alonewolf @witcheer @UnslothAI @LottoLabs

You can now fine-tune Qwen3.5 just need 5GB VRAM to train Qwen3.5-2B LoRA locally, 1.5x faster with 50% less VRAM. Qwen3.5-4B Colab: - colab.research.google.com/github/unsloth… - github.com/unslothai/unsl…

SUPER GEMMA 4 26B UNCENSORED GGUF v2 IS INSANE, - 0/100 refusals (actually uncensored) - Fixed all the tool-call + tokenizer jank - 90% faster prompt processing - Sharper, smarter, way more capable responses - Perfect local beast for llama.cpp Runs on 16 GB ,18, 22 GB VRAM (16.8 GB Q4_K_M file) - huggingface.co/Jiunsong/super…

Ever wondered how a 35B parameter model can run on consumer GPUs? Qwen3.6-35B-A3B-NVFP4 is a Mixture of Experts model quantized to FP4, packing huge intelligence into a tiny memory footprint. This is the future of efficient AI.

Hope you enjoy this!

A satellite has been jamming GPS over Europe boingboing.net/2026/06/05/a-s…

It's alive!

I've had this side project on the backburner for a long time... The XGecu line of universal programmers are awesome, but they come with some of the sketchiest proprietary Windows software ever. (XGPro) MiniPro is an software program made to interact with the XGecu hardware. An open source XGPro alternative.

My First RCE by Reverse Engineering an EXE File With the Help of AI A secure web app → a JS file leaking a download endpoint → a .NET binary → AI-assisted reverse engineering → a localhost WebSocket with no origin check → RCE Write-up: blog.voorivex.team/first-rce-via-…

i just ran Google's brand new Unsloth Gemma4 12B dense GGUF on my RTX 4060 using llama.cpp + CUDA 13.2 21 tokens per second. on a budget consumer GPU. locally. no API. no cloud. no subscription. and the benchmarks are absolutely cooked # first let's talk architecture because this is genuinely different every multimodal model you've used has a frozen vision encoder + frozen audio encoder + LLM backbone glued together Gemma 4 12B is different it's a single decoder only transformer. that's it. vision? raw 48×48 pixel patches → one matmul → projected directly into the LLM audio? raw 16kHz signal sliced into 40ms frames → linear projection → same LLM input space no encoder tax. no latency penalty. no fragmented memory to put the encoder savings in perspective: old Gemma 4 26B approach: - 550M param vision encoder (frozen) - 300M param audio encoder (frozen) - LLM backbone Gemma 4 12B: - 35M param vision embedder (a single matmul) - no audio encoder at all - LLM backbone handles EVERYTHING 550M → 35M for vision alone. that's a 15x reduction this is why the gemma-4-12b-it-Q4_K_M.gguf is just 6.6 GBs!!! and it has 256K native context context # Benchmarks: AIME 2026 (math olympiad): 77.5% GPQA Diamond (expert science): 78.8% LiveCodeBench v6 (real code): 72% Codeforces ELO: 1659 MMLU Pro: 77.2% MATH-Vision: 79.7% BigBench Extra Hard: 53% inference → llama.cpp, LM Studio, vLLM, SGLang llamacpp flags: -m "gemma-4-12b-it-Q4_K_M.gguf" -ngl 99 -c 8000 -v --port 8080 Available on huggingface now! Link below

Google Gemma @googlegemma

2 weeks ago

Meet Gemma 4 12B! A unified, encoder-free multimodal model designed to bring high-performance intelligence directly to your laptop, and released under an Apache 2.0 license. Bridging the gap between edge efficiency and advanced reasoning. Here is what’s new with Gemma 4 12B: 👇

403 2K 12K 3.2M 5K

Your government can't see what you're doing online. obfs4 bridges + ProxyChains = encrypted noise. No signatures. No flags. No trace. New identity every 30 seconds. This is why they want to ban encryption. 🧅 #Tor #ProxyChains #OPSEC #Cybersecurity #InfoSec

Spent 4 months building the ultimate mosquito killer: an artillery cannon guided by computer vision + deep learning. Trained a custom model to detect and lock onto mosquitoes using a DSLR + zoom lens setup. The dataset collection phase was brutal — the mosquitoes definitely fought back 🦟

Discovers real IPs behind Cloudflare github.com/musana/CF-Hero