DeepSeek is currently training a new model with 2 trillion parameters, with plans to eventually build an 8 trillion parameter model.
Chinese regulators are investigating DeepSeek and Moonshot AI regarding potential data leaks to the US-based company Anthropic. The probe follows rumors concerning Kimi executives, though the situation is currently described as an investigation rather than confirmed arrests.
Researchers introduce CodeMidas, an agentic pipeline that constructs reinforcement learning environments from implemented functionality in open-source codebases, using source code as the sole input. The method allocates agentic compute to explore functionality, construct execution-grounded tests, and validate tasks through repeated rollouts, resulting in a dataset of 5,545 training tasks across 23 programming languages. Training MiMo-V2.5 on these tasks with GRPO improves performance on five diverse benchmarks, including DeepSWE (+11.7%), ProgramBench (+17%), and Terminal-Bench v2.1 (+8.5%). Trajectory analysis indicates the RL-trained agent exhibits better behaviors such as increased codebase exploration and more diverse self-verification. These results establish source code as a scalable foundation for constructing RL environments that improve coding agents across diverse software tasks.
DeepSeek has released DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts model designed to address the high costs of long-context computation and large KV caches in agentic workloads.
Atria Dawn Preview is a foundation agentic language model designed for scientific research and engineering workflows, trained via a Verifiable Experience Pipeline that connects tool-mediated interactions to executable environments. Across 16 benchmarks spanning real-world research, engineering, and digital work, the model is competitive with frontier agents and achieves the highest reported score on five of them.
DeepSeek V4.1 Flash has taken first place on the new Astra benchmark within the Artificial Analysis Intelligence Index v4.3 update.
We use cookies to measure traffic and improve the site. You can accept or decline analytics cookies. Privacy policy