AI Startups news and analysis on Web Pulse: 10805 curated articles tracking the latest AI Startups developments, tools, and research, updated continuously from vetted sources.
Reflection introduced Beam, a 501-billion-parameter mixture-of-experts model with 23 billion active parameters, on Monday, and Mistral said Tuesday it is finishing Mistral Large 4, a 1-trillion-parameter multimodal model…
Mistral unveiled a new AI model that the French company says rivals the best open systems from China, according to CNBC Technology. The release comes as Western developers race to compete with Chinese companies in open-w…
Mistral launched a public preview of Mistral Large 4 (ML4), a 1-trillion-parameter open-weight flagship model, on Tuesday, with weights to be released on October 27. Mistral says the model was trained on 4,000 NVIDIA Gra…
Reflection released Beam, its first open-weight model, a mixture-of-experts system that activates 23 billion of its 501 billion parameters per token and aims to match GLM 5.2 on coding and reasoning while using three to …
China's CAC, NDRC and MIIT issued the Implementation Opinions on the Standardized Application and Innovative Development of Intelligent Agents on May 8, 2026, enforceable since July 15, 2026, which the article calls the …
Cohere launched North 2, an enterprise AI platform that orchestrates autonomous agents across multi-step workflows, retains context between sessions, and runs on-premises, in the cloud, or fully air-gapped. North 2 is mo…
Reflection AI, a New York startup founded in 2024 by former Google DeepMind researchers Misha Laskin and Ioannis Antonoglou, unveiled Beam on Monday, an open-weight text model with 501 billion total parameters that activ…
Reflection announced Beam, a text-only 501B-total / 23B-active mixture-of-experts model for coding, agentic and scientific work, trained from scratch on 23.8T pretraining tokens with full weights under Apache 2.0 due thi…
Reflection AI, an AI startup founded by two former Google DeepMind researchers and backed by Nvidia, unveiled an open-weight model called Beam on Monday (Oct 5) that it says scores comparably on advanced reasoning benchm…
Reflection AI Inc. introduced Beam, an open-source large language model with 501 billion parameters, which the company says matches or beats GLM-5.2 on some tasks while using between one third and one fourth the hardware…
Reflection AI unveiled Beam, an open-weight AI model with 501 billion total parameters and 23 billion active per token, on October 5, 2026, with full weights and technical documentation slated for release later in Octobe…
Reflection AI unveiled Beam on October 5, its first frontier open-weight model, which the Nvidia-backed startup says matches the reasoning performance of Z.ai's GLM-5.2 while being three to four times more efficient than…
Reflection AI, a New York-based startup founded in 2024 by former DeepMind researchers, released its first open-weight model, Beam, which CEO Misha Laskin said needs three to four times less computing power to reason thr…
Reflection AI published benchmark scores for Beam, its first open-weight model, showing 80.1 on Terminal-Bench 2.1 versus Z.ai's GLM 5.2 at 81.0 and 65.5 on SWE-bench Pro v1 versus GLM 5.2's 62.1, ahead of the model's pl…
Reflection, a startup backed by Nvidia, Sequoia, and Lightspeed and last valued at $25 billion, is releasing its first open-weight model, Beam, a 501-billion-parameter model trained from scratch under an Apache 2.0 licen…
Reflection AI, a New York City-based startup founded in March 2024 by former Google DeepMind researchers Misha Laskin and Ioannis Antonoglou, unveiled Beam, its first open-source model, which the company says matches Bei…
Synopsys struck a revenue-sharing deal with OpenAI to co-develop GPT-Synopsys, an AI model trained and tuned for chip design tasks, Synopsys said at its Investor Day in New York City, as reported by Reuters on September …
Amazon Web Services launched a public preview of Amazon Bedrock Managed Agents powered by OpenAI, a service that lets enterprises build and run AI agents on OpenAI models entirely inside their own AWS environments using …
HIVE Digital Technologies ran large language model pre-training for Columbia University's Department of Industrial Engineering and Operations Research on Nvidia A40 GPUs located in Asunción, Paraguay, operated remotely f…
Lola Vision Systems, a fabless chip startup founded in 2023 and based in the Washington, DC and Los Angeles areas, is developing the LVS-250, a five-chiplet edge AI processor it says delivers 230 TOPS at INT8 precision w…