Foundation Model

Google Launches Gemma 3: Top Single-GPU Multimodal Model for Mobile, 27B Beats o3-mini

Google Gemma 3: Open-source, multimodal AI with 128k context, 27B model outperforms o3-mini. Runs on single GPU/TPU, supports 140+ languages, and excels…

Mar 12 •

Alibaba Releases R1-Omni: First Full-Modality Emotion Recognition with DeepSeek-Style RLVR

Discover R1-Omni: Alibaba's open-source full-modality LLM that integrates DeepSeek-style RLVR for enhanced emotion recognition across video, audio, and…

Mar 11 •

Microsoft Phi-4 Family: 5.6B Multimodal Model Beats GPT-4o, 3.8B Competes with Qwen-7B

Explore Microsoft's Phi-4 models: Phi-4-multimodal integrates speech, vision, and text, while Phi-4-mini delivers impressive reasoning and performance…

Feb 27 •

Claude 3.7 Sonnet: The First Hybrid Reasoning Model Dominating Real Coding Tasks

Claude 3.7 Sonnet: The world's first hybrid reasoning model, excelling in coding and real-world tasks, with powerful features like intelligent coding…

Feb 25 •

Muon Optimizer: 48% Less Compute Than AdamW, Compatible with DeepSeek

Discover how Dark Side of the Moon's improved Muon optimizer cuts computational requirements by 48% over AdamW, scales to larger models, and is…

Feb 23 •

Microsoft Releases BioEmu-1: A Nobel-Worthy Breakthrough with Potential to Cure Cancer

Microsoft's BioEmu-1 is a breakthrough protein generation model, offering ultra-efficient predictions to accelerate cancer treatment design with…

Feb 22 •

Grok3 Free Trial vs DeepSeek & ChatGPT: Real-World Comparison

Grok-3, the world's strongest large model, now offers a free trial. Explore its powerful features, including advanced coding, reasoning, and deep…

Feb 20 •

DeepSeek Launches CODEI/O: Enhancing Large Model Inference with Thought Chains

DeepSeek's CODEI/O dataset enhances model reasoning by transforming code into natural language thought chains, improving performance across various…

Feb 17 •

Claude 4 & Thinking Model: Step-by-Step Reasoning and Controllable Budget

Claude 4 combines reasoning and general models, offering controllable token budgets and step-by-step reasoning for complex tasks. Get ready for the next…

Feb 16 •

Perplexity Launches Deep Research for Free: Outperforms R1, o3-mini, and More

Perplexity launches Deep Research, offering free, fast, expert-level reports on various topics. Outperforms models like R1 and o3-mini, with high…

Feb 15 •

DeepSeek R1: Auto-generates GPU Kernels, Shocking NVIDIA

Discover how NVIDIA is leveraging DeepSeek-R1's inference-time scaling to automatically generate optimized GPU kernels, surpassing skilled engineers and…

Feb 13 •

DeepSeek's GitHub Stars Surpass OpenAI in Just Two Months

DeepSeek surpasses OpenAI in GitHub Stars! With its MoE-powered V3 and reinforcement learning-driven R1, DeepSeek is redefining open-source AI…

Feb 7 •

#nojs-banner { position: fixed; bottom: 0; left: 0; padding: 16px 16px 16px 32px; width: 100%; box-sizing: border-box; background: red; color: white; font-family: -apple-system, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; font-size: 13px; line-height: 13px; } #nojs-banner a { color: inherit; text-decoration: underline; } This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts