Search AI Products and News

The Silicon Base Flow platform has launched the open-source MoE model Kimi K2 developed by Moonshot AI. The model has a total of 1T parameters and 32B activated parameters, supports a context length of 128K, and performs excellently in coding and agent tasks. The pricing is 4 yuan per million tokens for input and 16 yuan per million tokens for output. New users can get 14 yuan in trial credit upon registration. The model has three technical advantages: 15.5T tokens of large-scale training, MuonClip optimizer for stable expansion, and design optimized for agent tasks. Tests show that it excels in coding

Liquid AI opensources the next-generation edge AI model LFM2, available in three versions with 350M to 1.2B parameters. The model features an innovative architecture, achieving twice the inference speed and three times the training efficiency on edge devices, supporting 32K long context processing. LFM2 performs exceptionally well in tasks such as instruction following, outperforming models of similar scale, making it particularly suitable for privacy-sensitive scenarios. Fully open-sourced through Hugging Face, this marks the first time a U.S. company has surpassed Chinese open-source models in the field of efficient small models. Liquid AI

SpaceX plans to invest $2B in Musk's AI firm xAI, marking its first direct investment. This aligns with xAI's $5B equity funding round. SpaceX uses xAI's chatbot Grok for Starlink and aims to deepen collaboration. Despite Grok's controversies, Tesla plans to integrate it, showcasing Musk's strategy of inter-company synergy to advance AI.....

The CEO of American AI company Perplexity publicly praised the newly released Kimi K2 model with one trillion parameters from Moonshot AI, stating that they may conduct training based on this model. Kimi K2 uses a mixture-of-experts architecture, showing excellent performance in tasks such as code generation, intelligent agents, and mathematical reasoning, and has achieved leading results in multiple benchmark tests. Perplexity previously used the DeepSeek R1 model, and this statement shows that they are actively seeking technological collaborations to drive AI product development. Kimi K

IndexTTS2 is an upcoming film-grade TTS model with 3 breakthroughs: 1) Local deployment & open weights; 2) Zero-shot voice cloning; 3) Pioneering emotion cloning. Supports precise duration control for dubbing (Chinese/English).....

MoonDark releases trillion-parameter open-source model Kimi K2, featuring MoE with 32B active params. Excels in coding/math, supports tool use & code execution. Uses MuonClip optimizer, offers API & base model. Key competition to closed-source AI, available via official platforms.....

SpaceX invests $2B in xAI to compete with OpenAI. xAI, valued at $113B, integrates Grok AI in Starlink support and Tesla's Optimus. Despite past issues, Grok4 offers 100x more power, running on a supercomputer with 100K H100 GPUs. Musk calls it the 'smartest AI'.....

Moonshot AI officially released its latest creation - the Kimi K2 model, and simultaneously announced its open source. This foundation model based on the MoE architecture has gained widespread attention in the AI field since its release, thanks to its strong coding capabilities and excellent general Agent task processing abilities. The Kimi K2 model has a total of 1T parameters, with 32B activated parameters. It has achieved top performance among open-source models in a series of benchmark performance tests such as SWE Bench Verified, Tau2, and AceBench.

Ziyuan's robot Lingxi X2-N features dual-mode design: wheeled for mobility and legged for obstacle-crossing, carrying 6kg. It adapts to complex terrains with excellent balance and load capacity.....

On July 4, 2025, Kunlun Xiwang continued to open-source the second-generation reward model Skywork-Reward-V2 series. This series includes 8 reward models based on different foundation models, with parameter sizes ranging from 600 million to 8 billion. Upon its release, it won all seven major reward model evaluation rankings, becoming a focus in the open-source reward model field. Reward models play a key role in the reinforcement learning from human feedback (RLHF) process. To build the next generation of reward models, Kunlun Xiwang has constructed a dataset containing 40 million

Large Language Models (LLMs) have achieved significant progress in complex reasoning tasks by combining task prompts with large-scale reinforcement learning (RL), as demonstrated by models like Deepseek-R1-Zero, which directly apply reinforcement learning to base models, showcasing strong reasoning capabilities. However, this success is difficult to replicate across different base model families, especially within the Llama series. This raises a core question: what factors lead to inconsistent performance of different base models during reinforcement learning? How does reinforcement learning perform in

In the media Q&A session after today's Honor Magic V5 and AI Terminal Ecosystem Launch, Honor CEO Li Jian, CFO Peng Qiuen, and Product Line President Fang Fei had in-depth discussions with the media. During the event, Honor officially announced support for the MCP and A2A protocols, and revealed that it will collaborate deeply with partners such as Alibaba, BYD, and Midea in the fields of intelligent service ecosystem, smart vehicle networking, and smart home. Honor CEO Li Jian emphasized in the conversation that 'openness' is the core philosophy of Honor. He pointed out...

Artificial intelligence once again stirs up the field of drug development! Chai Discovery recently launched a new AI model called Chai-2, which has drawn widespread attention with its breakthrough technology in molecular design. Chai-2 achieves zero-shot antibody design with a success rate of 16%-20%, hundreds of times higher than traditional methods, shortening the drug development cycle from months or even years to just two weeks. Zero-shot antibody design breaks through traditional bottlenecks. Chai-2 is a multi-modal generative AI model developed by Chai Discovery, specifically designed for...

Recently, Meta Platforms is facing a major decision, possibly abandoning its self-developed Llama AI model and adopting artificial intelligence systems from competitors such as OpenAI and Anthropic. This change reflects a significant adjustment in Meta's open-source AI strategy and also shows the company's dissatisfaction with its own product performance. The turning point occurred at the Llama 4 launch event in April, where the product was presented at Meta's Llama

Amid the rapid development of the global artificial intelligence industry, the US AI startup Anthropic officially announced on June 24 that it will open its first Asia-Pacific office in Tokyo this fall. This news undoubtedly injects new vitality into the AI ecosystem in Japan and the entire Asia-Pacific region. Founded in 2020, Anthropic is committed to developing AI technologies centered around human needs, with the philosophy of building safe and controllable AI systems to better serve society through technology. The company is leading in this field.