Anthropic releases Claude Opus 4.1
Anthropic releases Claude Opus 4.1, an incremental closed-model update tuned for complex mathematics and long reasoning chains.
SEARCH
Search English reports
English reports are listed here. AI-generated drafts are clearly labeled until an editor reviews them.
AI-generated · reviewedImpact 45/100
Z.ai releases lightweight MoE model GLM-4.5 Air
Z.ai releases GLM-4.5 Air, a lighter 106-billion-parameter MoE variant optimized to reduce redundancy and improve inference speed.
Z.ai releases 355B MoE model GLM-4.5
Z.ai releases GLM-4.5, a 355-billion-parameter MoE model with updated expert routing and improved long-form structure generation.
Alibaba Cloud releases 480B MoE coding model Qwen3-Coder
Alibaba Cloud releases Qwen3-Coder, a 480-billion-parameter MoE model for complex software engineering and long-form code generation.
Google releases stable Gemini 2.5 Flash-Lite
Google releases the stable Gemini 2.5 Flash-Lite model for high-throughput workloads that prioritize low cost and low latency.
OpenAI merges Operator and deep research into ChatGPT agent
ChatGPT agent unifies web operation, deep research, code execution, and editable deliverables into a single agent.
Moonshot AI releases Kimi K2
Moonshot AI's trillion-parameter MoE model, continuing its emphasis on lossless recall across a very long context window.
xAI releases Grok 4
xAI releases Grok 4, its fourth-generation closed language model, with stronger cross-domain generalization and real-time web search retrieval.
Mistral AI releases Magistral Medium 1.1
An iteration of Magistral Medium with stronger handling and association across complex contexts.
Liquid AI releases LFM2
Liquid AI's second-generation lightweight model built on a dynamic liquid neural-network architecture.
Mistral AI releases Magistral Small 1.1
A small iterative update to the Magistral Small line that fine-tunes instruction-following accuracy.
Baidu releases ERNIE 4.5
An upgrade to Baidu's ERNIE series adopting an MoE architecture for better inference efficiency and generation quality.
Tencent releases Hunyuan A13B
Tencent released an open-source, medium-scale MoE model that balances performance with low deployment costs.
OpenAI releases GPT-o3 Pro
An advanced commercial variant that pairs systematic deep reasoning with high-precision text generation.
Anthropic makes Claude Code generally available
Alongside the Claude 4 launch, Anthropic moved Claude Code out of research preview to general availability across terminal, IDE, background tasks, and SDK.
Anthropic releases Claude 4 Opus
Anthropic's fourth-generation top-tier flagship, raising the ceiling on multimodal understanding and logical tasks.
Anthropic releases Claude 4 Sonnet
The mid-size member of the Claude 4 family, striking a better balance between response speed and reasoning.
Google releases Veo 3
Google released Veo 3, bringing native generation of ambient sound, sound effects, and character dialogue to its flagship video model.
OpenAI releases the Codex software-engineering agent
OpenAI released Codex, a software-engineering agent that handles coding tasks in parallel inside a cloud sandbox.
Alibaba releases Qwen3 with hybrid thinking modes
Qwen3 combines fast direct answers and longer reasoning in one open-weight family, spanning dense and MoE models up to a 235B flagship.