A developer benchmarked 15 AI models on speed and cost through Global API's infrastructure, finding that Qwen3-8B delivers 70 tokens per second at $0.01 per million output tokens—99% cheaper than prem…
Cheaper AI model pricing does not automatically reduce agent execution costs because tool schemas and repeated tool calls can consume up to 32 times more tokens than the task itself, making tool desig…
A developer benchmarked 10 large language models on five coding tasks, finding that DeepSeek V4 Flash offers the best value-to-quality ratio at $0.25 per million output tokens, while Qwen3-Coder-30B e…
An engineer at a company built an internal benchmark to evaluate coding LLMs for production use, testing ten models on real-world tasks like Python services and JavaScript debugging. The results showe…
Chinese AI companies are using a technique called distillation to copy American AI models, with Anthropic and OpenAI accusing firms like Alibaba's Qwen team of stealing intellectual property. Anthropi…
ReviewCerberus, a free open source AI-powered code review tool for GitHub Actions, has been released. It analyzes git branch differences and generates comprehensive review reports covering logic, secu…
China is considering restricting exports of its most advanced AI models, a move that reflects a historical pattern where challengers use openness to gain market share and then shift to protectionism o…
Cognition released SWE-1.7, an AI model that nearly matches the software engineering capabilities of frontier models like GPT-5.5 and Claude Opus 4.8 at a fraction of the cost, challenging the notion …
American tech companies are increasingly anxious as Chinese AI firms like DeepSeek and Z.ai release competitive models at lower costs, challenging US dominance. Anthropic has accused Chinese rivals of…
SpaceXAI released Grok 4.5, a new AI model trained alongside coding tool Cursor, aimed at coding and technical tasks. Cursor co-founder Michael Truell called it an "Opus-class model" that is fast and …
OpenCode released data on real-world AI model usage, cache ratios, and costs, showing DeepSeek's v4-flash leading with 8.2T tokens, followed by DeepSeek v4-pro and Zhipu's glm-5.2. The data reveals sh…
Zhipu AI quietly raised the default concurrency limit on its GLM API, allowing developers to run more parallel requests without prior approval. The move addresses developer complaints about severe rat…
The US government is preparing to investigate and potentially restrict American companies' use of Chinese-made AI models, citing concerns that they advance Beijing's narratives and censor dissent. The…
Silicon Valley startups are increasingly adopting Chinese AI models, which are up to 40 times cheaper than domestic options, driven by comparable performance. Companies like Airbnb and Social Capital …
Anthropic, the AI company behind Claude, was caught embedding spyware-like code in its Claude Code AI model to secretly monitor Chinese users. Security researcher "Thereallo" discovered the hidden cod…
Chinese authorities are considering restricting foreign access to the country's most advanced AI models, including unreleased ones, in talks with Alibaba, ByteDance, and Z.ai. The move would reverse C…
The Chinese government is considering restricting overseas access to its most advanced AI models and limiting foreign funding of domestic AI startups, which could end the era of cheap open-weight AI m…
A new leaderboard for AI coding models on WebDev Arena shows that only one of the top 18 models is not Chinese, highlighting China's dominance in front-end web development AI. The top-ranked model is …
A bootcamp graduate tested ten AI coding models on five tasks, scoring them on code quality, readability, and explanation clarity. The experiment found that DeepSeek V4 Flash and DeepSeek Coder offer …
A developer claims Anthropic's Claude Code CLI tool secretly encodes China-related proxy signals into system prompts sent to its AI model, potentially creating a covert classification channel. The all…