It hasn't had its “DeepSeek moment” yet, even though it's hands down China's best model now Probably due to hype around Grok But based on my TL it looks like people are finally starting to take notice (DeepSeek took a few days before it hit mainstream)
GENERATIVE AI
-
Kimi K2 Surpasses GPT-4.1 and Claude Opus on Benchmarks
By
–
For those unfamiliar with Kimi K2: – Surpasses models like GPT-4.1 and Claude 4 Opus on coding benchmarks
– Scores new highs on math and STEM tests among non-reasoning systems
– Doesn’t even have multimodal or reasoning capabilities yet kimi [dot] com -
AI-Generated Writing Reaches Tipping Point with China’s Kimi K2
By
–
We're at the tipping point for AI-generated writing It's been notoriously bad But China's Kimi K2, *an open-weight model*, is now topping creative writing benchmarks …without any reasoning capabilities
-
Microsoft Releases Phi-4-mini-flash-reasoning Model
By
–
Microsoft dropped Phi-4-mini-flash-reasoning on Hugging Face Built on a new hybrid architecture, the 3.8B model surpasses Phi-4-mini-Reasoning on major reasoning tasks, while delivering up to 10× higher throughput at 32K generation length
-
Meta Acquires PlayAI, Integrates Team into Superintelligence Lab
By
–
Meta has reportedly acquired voice AI startup PlayAI
— The Rundown AI (@TheRundownAI) 14 juillet 2025
The entire team of the company will now join the tech giant, working in its new Superintelligence Lab
They will be reporting to Johan Schalkwyk, who recently joined from Sesame AIpic.twitter.com/iorcXxRy5hMeta has reportedly acquired voice AI startup PlayAI The entire team of the company will now join the tech giant, working in its new Superintelligence Lab They will be reporting to Johan Schalkwyk, who recently joined from Sesame AI
-
OpenAI Delays Open-Weight Model Release for Safety Testing
By
–
OpenAI is delaying the release of its highly anticipated open-weight model CEO Sam Altman shared on X that the company needs time to "run additional safety tests and review high-risk areas" The model was originally planned for this week
-
Moonshot AI Releases Kimi-K2: 1T Parameter Open-Source MoE Model
By
–
Moonshot AI released Kimi-K2, a 1T param open-source MoE model that: —Surpasses frontier AI like GPT-4.1 and Claude Opus 4 on coding, with new highs on math and STEM
—Excels at agentic workflows with tool use
—Doesn't have multimodality or reasoning yet -
WordPress Categories for AI Topics
By
–
CI Agent committed 19 hours ago +1 -0 lines changed
grok4_system_turn_prompt_v8.j2 -
Do LLMs Really Reason or Just Simulate Reasoning?
By
–
Your poster says that LLM reasoning is fake by definition.
-
Reinforcement Learning of Large Language Models Course
By
–
Reinforcement Learning of Large Language Models Youtube playlist: https://
youtube.com/playlist?list=
PLir0BWtR5vRp5dqaouyMU-oTSzaU5LK9r
… Website: https://
ernestryu.com/courses/RL-LLM
.html
…
Tweet: