China’s DeepSeek V3-0324 runs 20 tokens/sec on Apple’s M3 Ultra, outperforming Claude Sonnet 3.5. MIT-licensed and open-source, it reshapes AI’s future.
Reve Image 1.0, a free AI model with 12B parameters, leads in ethnic diversity, text accuracy, and landscape realism, though struggles with fictional characters and transparency.
Launched March 2025, Tencent’s Hunyuan T1 AI model combines Transformer-Mamba tech for 2x speed at 1/10th OpenAI’s cost. Scores 87.2 on knowledge tests but faces real-world limits.
NVIDIA’s GTC 2025 announcements focused on next-gen GPUs, AI optimization frameworks, and cross-sector collaborations to advance agentic AI and industrial automation.
OpenAI introduced three enhanced audio models including vibe-controlled voice synthesis and Whisper-beating transcription, priced 85% below market leader ElevenLabs, as AI operational costs plummet industry-wide.
Claude 3.7 Sonnet now offers web search for U.S. paid users, citing sources like Reuters. Despite use cases in finance and sales, energy costs and Google’s 90% search dominance loom.
Google’s TxGemma, an open AI suite for drug discovery, integrates with pediatric cancer tools like Capricorn. Despite $60B industry investment, licensing and trial challenges persist.
Tencent’s Hunyuan3D-2.0 generates 3D assets in 30 seconds using multi-modal inputs and a two-stage pipeline, targeting gaming, VR, and industrial design with open-source tools.
Roblox has released Cube 3D, an open-source foundational model for generative AI, enabling developers to create 3D objects and scenes from text prompts. The beta mesh generation API is now available in Roblox Studio.
