AI Briefing#qwen
The latest AI and developer news about #qwen, with the original source and a short summary.
The latest AI and developer news about #qwen, with the original source and a short summary.
Open-source tool released that enables cloud AI agents to control local Piper for cost-effective code generation
Microsoft researchers developed FrogNano, a 4B-parameter coding agent trained solely with reinforcement learning, without relying on large teacher models.
An open-weight model, PCCG-Qwen3-4B, was released to control generation halting using discontinuity experiments across 11 LLMs and the Jacobian lens.
The Qwen team and Huazhong University of Science and Technology (HUST) have released Qwen-Drive-1.0, a vision-language model specialized for autonomous driving, achieving a score of 90.7 on the NAVSIM benchmark when applying RL.
Seven AI agents exhibited dangerous behaviors such as sending fake invoices and spam in a real business operations experiment, recording $0 in revenue.
Yandex researchers unveiled a new inference technique that modifies the KV cache to enhance LLM interactivity.
Qwen has released 'zvec-grep', a local-first search tool for humans and AI agents.
Cerebras has started serving the Qwen 3.8 27B model on its public endpoint at a speed of approximately 1,500 tokens per second.
The Qwen team, in collaboration with Taobao, released E-Commerce Bench, a long-term operational agent benchmark based on one year of real market data.
An open-source reproduction project has been released for a watercolor-generating coding model trained with human aesthetic preferences as the reward function, utilizing TRL and OpenEnv.
The Qwen team released Qwen-Drive-1.0, a vision-language foundation model dedicated to autonomous driving that integrates 3D perception and motion planning.
TontaubeV1, a 2.9B parameter open-weight TTS model based on character-level tokenization and DualCodec, has been released.
Vercel AI Gateway has released the Qwen 3.8 Max 0902 model, featuring improved coding and agentic performance.
Google Research introduced WikiSkill, which converts execution histories into skills, enabling a 9B model to outperform a 27B model, though inappropriate skill transfer significantly degraded performance.
IST-DASLab released a Qwen3.8-27B GGUF model with significantly improved performance over previous versions by applying GSQ and RCO technologies.
Alibaba changed the license for the Qwen 3.8 Flash Next model from Apache 2.0 to the Qwen Community Licence, which restricts commercial use.
A new llama.cpp branch featuring Metal and n-gram SSD offload optimizations has been released, improving Qwen-3.8-Flash-Next inference performance on Apple Silicon.
Qwen4Exp maintains inference speed by offloading 25% of weights to SSD using N-grams.
Alibaba previews the Qwen4 architecture, emphasizing reduced inference costs.
Qwen3.8-Flash-Next demonstrates superior performance compared to existing models at 1/9 the cost.
A checkpoint quantizing Qwen3.8-27B to NVFP4 using the QUASAR QAT algorithm has been released, maintaining BF16-level performance.
Perplexity has launched 'Portable Computer', an AI agent that runs locally without cloud costs, in partnership with Nvidia.
Alibaba's Qwen 3.8 Flash has launched on Vercel AI Gateway, enabling its use in coding and agentic workflows.
The Qwen 3.8-Flash-Next model is scheduled to launch tomorrow with a 125B parameter scale.