The Vision Encoder: Our Secret Weapon 🔥 An ultra-accurate... vision encoder plate, combined with 5μm-resolution optical feedback, calibrates the H2D’s motion system to achieve consistent 50μm accuracy across the entire build plate. #BambuH2Dshow more

This optical device adds accuracy and enables those with... poor vision to compete.show more

the encoder

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 belowshow more

Absolute Encoder Vs Incremental Encoder

JUAN SOTO WITH AN ABSOLUTE LASER TO GET THE... RUNNER AT THE PLATE! 🔥show more

Stepped to the plate with our torpedo Bratt.

Crazy how transformers work! It’s the encoder part

Can a VLM see without a vision encoder? We... trained one for $100, inspired by Gemma 4 12B. Latency on an M3 Pro MacBook: 112 ms -> 1.1 ms for the image path 30% lower end-to-end image+LLM The architecture is just: patchify the image -> linear projection with pos embeddings -> LLM Writeup:show more

🚨 Only on 365Scores: Exclusive Live Match Tracking 🚨... We’ve officially levelled up. Experience football like never before with our Exclusive Live Match Tracker, now powered by industry-leading Computer Vision technology. What makes it a game-changer? 🤖 Computer Vision Precision: Our AI-driven tech tracks every play with pinpoint accuracy. 🏃‍♂️💨 Fluid Motion: See more accurate and realistic player movement across the pitch. ⚡️Enhanced Visuals: A premium, high-tech interpretation of every match moment. 📲 Download the app now to witness the tech in action! 😎 📺 Available currently in UEFA Competitions 🇪🇺, Bundesliga and Bundesliga 2 🇩🇪show more

NO LIDAR, NO RADAR – TESLA BETS EVERYTHING ON... VISION Tesla’s FSD cameras called the future early – just eyes on the road, silicon style! Vision Facts: •⁠ ⁠GPS + passive optical, nothing else in the stack •⁠ ⁠Road system built for cameras, not sci-fi sensors •⁠ ⁠Superhuman driving is possible with vision alone Tesla called it in 2017 – solve vision, solve autonomy! Source: Elon Musk, TED Talks, 2017show more

Experimenting with Glassmorphism and hand interactions on the Apple... Vision Pro. Reflections are the Apple Vision hallmark signature. But how can we achieve refractions? Made with RealityKitshow more

The plate has no nutritional value, it's just an... empty decorative plateshow more

Build the team, present the vision, kick ass.

I shouldn't be allowed to use Flipnote Encoder

The Bambu Lab Vision Plate actually works! Unfortunately, I... learned this the hard way after printing 500+ pieces wrong for GatlingTommy MOAB Story Time 😂🤘 #3DPrinting #Nerfshow more

manual control of the monster mwir lens achieved. gonna... hook into the magnetic end stop + encoder to replace the proprietary Telic Optics board.show more

Introducing ✨ Aya Vision ✨ - an open-weights model... to connect our world through language and vision Aya Vision adds breakthrough multimodal capabilities to our state-of-the-art multilingual 8B and 32B models. 🌿show more

The number plate 🔥🔥🔥

A teaser of the latest 3D DUSt3R based models... we’re presenting at @cvprconference Discover MUSt3R & Pow3R, universal encoder DUNE + research in navigation, vizloc, segmentation & human motion understanding! All our #CVPR2025 papers are here ➡️show more

Apple Vision Pro supports Direct touch input across the... system 🤩 #AppleVisionPro #visionOS #WWDC23 #SpatialComputing #AR #VR #MRshow more