A developer benchmarked ten AI coding models across five tasks and four languages to find the best value for production use. The results show DeepSeek V4 Flash and DeepSeek Coder offer the highest val…
Alibaba's Qwen3.8-27B, a 27-billion-parameter model, matched OpenAI's GPT-5.6 Luna on the Artificial Analysis Intelligence Index, outperforming larger rivals while running on everyday hardware. The mo…
Chinese AI developer Zhipu launched GLM-5.3, a coding-focused model that unexpectedly developed advanced vulnerability discovery capabilities, scoring 84.5% on the CyberGym benchmark, ahead of Anthrop…
A developer spent $47 testing nine multimodal AI APIs from Chinese labs, including Qwen, GLM, Hunyuan, and Doubao, to find cost-effective alternatives to OpenAI's GPT-4o for image-heavy workloads. The…
Chinese AI developer Zhipu has launched GLM-5.3, a coding-focused AI model that scored 84.5% on CyberGym, ahead of Anthropic's Mythos 5 (83.8%) and OpenAI's GPT-5.6 Sol (83.6%), but trails both on Exp…
Z.ai, the international brand of Chinese AI company Zhipu, launched GLM-5.3, an update focused on coding, long-horizon tasks, and cybersecurity, claiming a 50% higher score than GLM-5.2 on its interna…
Chinese AI company Zhipu launched GLM-5.3, claiming it outperforms Anthropic's Fable 5 and OpenAI's GPT-5.6 Sol on the CyberGym benchmark for vulnerability discovery, and found 2,436 vulnerabilities a…
Frontier AI models are activating far fewer parameters per token than three years ago — Zhipu's GLM-5.2 activates about 40 billion, Alibaba's Qwen3.5 runs 17 billion active, and DeepSeek V4-Flash gets…
Chinese AI firm Zhipu (Z.ai) launched its flagship GLM-5.3 model, claiming it outperformed Anthropic's Mythos 5 and OpenAI's GPT-5.6 Sol on the CyberGym benchmark with an 84.5% success rate, though it…
Chinese AI company Zhipu is approaching 7 million registered API users on its MaaS open platform, adding roughly 2 million since early July, and has activated more than 50,000 domestically developed A…
Zhipu's MaaS open platform has nearly 7 million registered API users, about 2 million more than in early July, and has activated more than 50,000 domestically developed AI chips to meet rising inferen…
OtterMind AI operates on a credit-based system with a free tier offering access to OtterMind Light, while paid users can use OtterMind Pro and frontier models from OpenAI, Anthropic, and Zhipu, all pr…
Kimi K3, a ~2.8T parameter open-weight coding model from Moonshot AI, ranks first on Frontend Code Arena with a score of 1,679, ahead of the closed Fable 5 at 1,631, and its weights were released on H…
A developer analyzed the costs of self-hosting open-weights AI models on AWS, finding that a team of 10 can run a strong model like Llama 4 Maverick for about $1,250 per month during business hours us…
LG AI Research released K-EXAONE 2.0, Korea's largest AI model with 750 billion parameters, to the open-source community Hugging Face. The model, intended for the Korean government's National AI Found…
DeepSeek founder Liang Wenfeng said during an investor call that the company has no KPIs, no written vision, and no formal organization, operating instead on goodwill and an obsession with AGI. He sta…
A developer built GrokRoute, a local proxy that routes requests from Grok Build to multiple AI providers including Agnes, Groq, Zhipu, Laguna, and Ollama with auto-fallback. The single-file Python pro…
The corporate fad of 'tokenmaxxing' — maximizing AI token usage as a productivity metric — is fading as workplaces face soaring costs without proportional gains, according to a report from the Associa…
Chinese AI developers may start charging cloud platforms commercial licensing fees to host their open-weight models, Goldman Sachs told the South China Morning Post, a shift that could alter the econo…
Chinese AI labs including DeepSeek, Alibaba's Qwen, Moonshot's Kimi, Zhipu's GLM, and MiniMax have shipped models that are 10–30x cheaper than Western flagships while scoring within a few points of th…