Markus Strazds PRO

MStrazds

2 54

AI & ML interests

None yet

Recent Activity

liked a model 9 days ago

empero-ai/Qwythos-9B-Claude-Mythos-5-1M-GGUF

reacted to Quazim0t0's post with ❤️ 9 days ago

🌼 DaisyChain-Web: train a language model with friends or by yourself with multiple devices, in the browser, no install Open a webpage, share a room link, and every device that joins becomes part of the training cluster. Phones, laptops, old PCs: they connect peer-to-peer over WebRTC and train one shared transformer together, entirely in the browser. What's actually happening under the hood: 🧠 A mini transformer LM trains on FineWeb-Edu, streamed live from the HuggingFace Hub. Each device pulls its own slice (data parallelism), tokenized with our 16.5k-token Spikewhale tokenizer ⚡ Every single multiply runs through verified INT8 neural units, no float fallback. On WebGPU browsers it uses the GPU's DP4A integer dot-product hardware, admitted only after proving bit-identical results against the verified units, with a 3×INT8 fast-accurate scheme (CUTLASS's 3xTF32 trick, ported to 8-bit) 🔒 Devices average gradients every step under a sync guard: a per-step roster protocol plus weight-hash verification keeps every device's model bit-identical. If anything drifts, training stops instead of silently forking 📊 Live logs show exactly what every device contributes, step by step 💾 When you're done: test generations right on the page, download a checkpoint, or grab the inference kit, a single self-contained HTML file with the weights baked in that runs generations offline, anywhere Works solo too. Every extra device just grows the effective batch. 👉 Try it: https://huggingface.co/spaces/Quazim0t0/DaisyChain-Web 🛠 Training framework: https://huggingface.co/DaisyChainAI/DaisyChain-Train Updates: - Block-scaled INT8 quantization - Batched attention GEMM - Fused dequant+ReLU epilogue - Weight-tied unembedding - WebSocket relay fallback - Server keepalive ping/pong every 30s - disconnected-state redial - Visibility/network-change reconnect

upvoted an article 28 days ago

Kog Laneformer 2B: The Latency-First Model Behind Kog Inference Engine

View all activity

Organizations

liked a model 9 days ago

empero-ai/Qwythos-9B-Claude-Mythos-5-1M-GGUF

Image-Text-to-Text • 9B • Updated 9 days ago • 2.13M • 2.42k

liked 3 models 3 months ago

liked 3 models 4 months ago

jinaai/jina-embeddings-v5-text-nano

Feature Extraction • 0.2B • Updated Apr 15 • 1.19M • 92

Qwen/Qwen3.5-9B

Image-Text-to-Text • 10B • Updated Mar 2 • 9.42M • • 1.75k

Qwen/Qwen3-Embedding-0.6B

Feature Extraction • 0.6B • Updated Apr 20 • 9.87M • • 1.12k

liked 2 models 5 months ago

nomic-ai/nomic-embed-text-v1.5

Qwen/Qwen3.5-35B-A3B

Image-Text-to-Text • 36B • Updated Apr 24 • 2.37M • • 1.47k

liked a Space 5 months ago

Invisible Watermark Against Unauthorized AI Training — Text, Image & Video Protection

⚡

One embed. Four invisible layers. 34 attacks defeated.

liked 6 models 5 months ago

unsloth/Qwen3.5-35B-A3B-GGUF

Image-Text-to-Text • 35B • Updated Mar 5 • 161k • 862

Nanbeige/Nanbeige4.1-3B

Text Generation • 4B • Updated Mar 25 • 18.3k • • 1.15k

Open-Orca/Mistral-7B-OpenOrca

Text Generation • Updated Nov 18, 2023 • 1.11k • • 689

tobil/qmd-query-expansion-1.7B

2B • Updated Feb 22 • 1.46k • 8

google/flan-t5-small

77M • Updated Oct 10, 2023 • 949k • 486

z-lab/Qwen3-4B-DFlash-b16

Text Generation • 0.5B • Updated Apr 7 • 23.7k • 29

liked 4 models 6 months ago

onnx-community/Qwen3-0.6B-DQ-ONNX

Text Generation • Updated Nov 20, 2025 • 53 • 3

Qwen/Qwen3-0.6B

Text Generation • 0.8B • Updated Jul 26, 2025 • 25.8M • • 1.44k

LiquidAI/LFM2.5-1.2B-Thinking-ONNX

Text Generation • Updated Mar 30 • 604 • 18

LiquidAI/LFM2.5-1.2B-Thinking

Text Generation • 1B • Updated Mar 30 • 18.8k • 383

Markus Strazds PRO

AI & ML interests

Recent Activity

Organizations

MStrazds's activity

Invisible Watermark Against Unauthorized AI Training — Text, Image & Video Protection