A backend engineer's field notes from 2026 reveal a 350× price spread between the cheapest and most expensive AI APIs, with Qwen3-8B at $0.01/M output tokens and DeepSeek V4 Pro at $3.50/M. The engine…
StepFun, a Chinese AI unicorn, is developing its first agentic AI smartphone, signaling a shift in AI competition from data centers to mobile devices. The company, backed by electronics supply chain f…
China will debut the world's first AI agent phone at the World Artificial Intelligence Conference in Shanghai from July 17-20, according to Shanghai official Tang Wenkan. Huawei will also showcase its…
Huawei will debut its Atlas 950 SuperPoD computing cluster and the world's first AI agent phone at the World Artificial Intelligence Conference in Shanghai from July 17-20. China's AI phone and comput…
A developer benchmarked 15 AI models through Global API's unified endpoint, measuring time to first token and tokens per second. Step-3.5-Flash from StepFun topped the speed leaderboard with 80 tokens…
AIWave has launched an API that provides access to over 50 Chinese AI models through a single OpenAI-compatible endpoint. The service supports models from DeepSeek, Zhipu, Qwen, and others, enabling d…
AIWave aggregates 50+ Chinese AI models behind a single OpenAI-compatible API endpoint, eliminating the need for multiple API keys, SDKs, and authentication schemes. Developers can switch between mode…
Step 3.7 Flash, released on May 29, 2026, is a structural upgrade to 3.5 Flash with a new vision encoder, runtime escalation, and a compute-control flag. The migration requires two environment variabl…
StepFun and MiniMax released new open-weight AI models over the weekend, with StepFun launching Step 3.7 Flash under a permissive Apache 2.0 license and MiniMax preparing to release the weights for it…
StepFun released Step 3.7 Flash, a 198-billion parameter sparse MoE model, which achieves 76.3% on SWE-Bench Verified, matching 97% of Claude Opus 4.6’s coding performance at one-ninth the cost—$0.19 …
StepFun released Step 3.7 Flash, a 198-billion parameter Mixture-of-Experts vision-language model with native image understanding and improved tool-use reliability for agentic coding and search workfl…
StepFun released Step 3.7 Flash, a 198B-parameter Mixture-of-Experts vision-language model with 11B active parameters, for enterprise deployment on NVIDIA-accelerated infrastructure. The multimodal AI…
Researchers analyzing weight files from over a dozen open-weight language models — spanning labs including Qwen, DeepSeek, Google, and OpenAI, from 0.6 billion to 1.4 trillion parameters — found that …