240ms TTFA, 2x real-time for @Alibaba_Qwen TTS 0.6B model running on iPhone 16e NPU. Soon in @TheStageAI SDK: github.com/TheStageAI/AppleS…

260 tok/s for @liquidai LFM2.5-230m running with @TheStageAI SDK on iPhone 16e NPU

Gemma4 E2B, compressed by @TheStageAI , from 9.3GB to 1.4GB, is running on iPhone 16e with tool calls! The smallest and the best quality checkpoints open-sourced! @GoogleDeepMind

Great communities make great products. At @TheStageAI, we’re building ANNA, our Autonomous Neural Networks Accelerator, for faster, cheaper inference. We need a Community Manager now. Be part of the early story →

How to measure the quality of text-to-image models? Our research team @TheStageAI put together a comprehensive guide to check perceptual quality, sharpness, color, prompt alignment, and more. All the tricky image quality questions researchers usually ask are covered here↓

🚀 Early access to ANNA: Automated NNs Accelerator now available! ✨ Get your access here: app.thestage.ai/contact Questions? DM or comment below! 💬 With ANNA, you can: 🔄 Simply upload your model, data, and desired metrics 🎛️ Fine-tune model size, latency, and quality with an intuitive slider 🔗 Combine multiple compression & acceleration algorithms in a single neural network ⚡ Boost performance by more than 2x with zero quality loss!

Quantization delivers speedup but can reduce quality. Our researchers prepared a tutorial showing how ANNA automatically quantizes Flux and accelerates it 2× while keeping quality high. Orig. model latency: 6.4 s. Check the link. DM or comment for early access.

Self-hosted text-to-image on H100 with @TheStageAI Elastic Models, accelerated from FLUX.1-schnell @bfl_ml. Our fastest model S generates a high-quality image in 0.5 s. Precompiled and ready-to-deploy – minimal cold start. Tutorial + access token inside if you want to try.

Can LLMs recognize ASCII art? Our tests show accelerated Elastic Models analyze line-by-line features and combine them using statistical patterns. Try it yourself with DeepSeek-Qwen-14B – 120 tok/s on H100, 40 tok/s on L40s, up to 3× faster. Free API token!

Our research team took @AIatMeta LLaMA-8B, quantized it with QLIP using post-training int8, applied SmoothQuant, and used pre-defined compiler-compatible NVIDIA configs. Why do this? Up to 2× fewer weights and 3.6× faster on one GPU. Try it with our simple Jupyter Notebook.

Our @TheStageAI team was happy to gain early access to the @nvidia B200 from @nebiusai and establish benchmarking for our optimized diffusion models. We now fully support inference of optimized models on B200 across various AI applications - LLMs, VLMs, Text-to-Image, Text-to-Video, Text-to-Music, Speech-to-text, and more! DM for access to our automated acceleration stack!

Meet Elastic MusicGen Large — our optimized fork of @metaai's MusicGen, powered by ANNA (@TheStageAI’s Automated Neural Network Accelerator): huggingface.co/TheStageAI/El… Ye @kanyewest used AI for vocals on "Bully," calling it the "next Auto-Tune." He switched up later, but tracks like "BEAUTY AND THE BEAST" keep the conversation buzzing. What's AI's true place in music's future? Here's @TheStageAI's take: — High-quality music from text prompts. — Pre-compiled for @NVIDIA H100/L40S. — Accelerates prototyping, custom styles, real-time. — 7M+ 20s tracks/month. — Full on-premise optimization & deployment. DM me for custom model optimization or other GPUs 😉

⌁ EUROPE SIGNAL: ACTIVE ⌁ ↳ Want to accelerate your model’s inference? ↳ These guys sure do. ✦ Berlin: mapped next steps with our investors Christophe Maire and Lukas Erbguth of Atlantic Labs. ✦ Paris: @NVIDIAGTC showed us what’s possible. ✦ Germany: more investor talks underway. The loop continues.

▚▞▚▞ DATA LOG: AI EUROPE ▚▞▚▞ For years, AI talk was all Silicon Valley. After @NVIDIA #GTCParis, one thing became clear: Europe’s AI ecosystem has already kicked into high gear. 🇫🇷 @MistralAI’s dropping open weights that actually run. 🇩🇪 @Aleph__Alpha building native multilingual models with explainability baked in. 🇩🇪 @DeepLcom quietly powering half the internet’s translations. 🇬🇧 @StabilityAI made open-source diffusion mainstream. 🇬🇧 @meetgranola is boosting productivity with discreet AI notepads. 🇫🇷 @OpenAI’s expanding in France, teaming up with local AI research and startups. European AI startups raised €3.4B in Q1 2025 — a 55% jump from last year, capturing 25% of global VC, per @dealroomco. There’s a lot happening under the radar — we couldn’t ignore it. We’re now helping early-stage teams across Europe get fast, reliable inference from day one. → More soon.

If you want to learn more about quantization of neural networks we have created educational podcast where you can dive deeper into efficient deployment of DNNs. First episode already here: youtube.com/watch?v=ip__DinD…