Moonshot AI released the weights for Kimi K3 on July 27, a 2.8-trillion-parameter Mixture-of-Experts model with a 1-million-token context window, scoring third globally behind Claude Fable 5 Max and G…
AI enthusiasts like Jeffrey Escobar, a 25-year-old from Miami, report feeling isolated as friends and family fail to share their passion for building with AI, with Escobar saying, "It's like I'm in a …
Moonshot AI released Kimi K3, an open-weights model with 2.78 trillion total parameters and 104.2 billion active per token, achieving a composite Intelligence Index score of 57 on Artificial Analysis,…
China's Moonshot AI is expected to close a financing round by the end of August at a valuation of up to US$50 billion, ahead of a potential Hong Kong IPO as early as year-end, according to sources. Th…
Moonshot AI, the Chinese developer of the Kimi chatbot, has reportedly opened a G-round pre-IPO financing at a $50 billion valuation, with investor interest intensifying after the launch of Kimi K3. A…
Data gravity is compounding integration debt in API-first AI stacks, making vendor lock-in a gradual process that most teams drift into rather than choose, according to a technical analysis by Glukhov…
Hardware wallet maker Coldcard reported that AI code review tools including Kimi K3, Claude Fable, and Codex 5.6 failed to detect the firmware vulnerability exploited in a Bitcoin theft that has compr…
At the Stanford AI Summit, OpenAI Chief People Officer Arvind KC found that nearly all attendees said AI made them more productive, but few were working less. Developer platform Kilo introduced 'Token…
Moonshot AI and KVCache.ai released AgentENV, a distributed sandbox runtime for AI agents that claims incremental snapshots in under 100 milliseconds, but a detailed analysis shows the complete fork-t…
A developer spent $300 to self-host Kimi K3, an inference model, and documented the experience, detailing the costs, performance, and practical challenges of running the model locally. The study provi…
A new GitHub project called Waste introduces a streaming approach to run massive AI models like the 2.78-trillion-parameter Kimi K3 in low-RAM environments by pulling activated weights directly from N…
Artificial Analysis launched its Endpoint Accuracy Index, measuring how much of an open weights model's accuracy each serverless API endpoint preserves, with initial coverage of GLM-5.2, gpt-oss-120b,…
In a benchmark run on Thursday, Fable and Sol tied at 33.3% (10/30 strict checkpoint passes) on SlopCodeBench, with Expo and Kimi K3 following at 26.7% and 23.3%, respectively. The evaluation, conduct…
Moonshot AI's Kimi K3, a 2.8 trillion parameter open-weight model, proved unexpectedly useful in seven real-world use cases after the author initially ignored it for two weeks, including debugging a m…
Moonshot AI released Kimi K3 in July 2026 as a 2.8T-parameter open-weight MoE model with a 1M token context window and native vision support, scoring 88.3 on Terminal-Bench 2.1, close to top closed mo…
Three independent benchmarks show that expensive AI defaults are rarely optimal: medium reasoning effort captures nearly all of Claude Opus 5's bug-fix gains, a three-model open-weight jury (GPT-OSS 1…
Nvidia commands over 75% of the global AI accelerator market as of mid-2026, with US entities controlling roughly 75% of global AI-ready GPU cluster performance, according to a report from Crypto Brie…
Chinese AI companies are intensifying competition with Silicon Valley, led by Moonshot AI's Kimi K3, a 2.8 trillion-parameter model that is the world's largest open AI model. The surge from firms like…
Alibaba Qwen announced Qwen3.8-Max, a 2.4T-parameter flagship model focused on coding, long-horizon agentic work, and multimodal reasoning, with open weights promised for next week alongside Qwen3.8-2…
Chinese AI startup Moonshot is reportedly seeking greater access to Nvidia's advanced Blackwell GPUs to develop its next-generation AI model, highlighting the growing importance of computing power in …