OpenAI

Latest AI news, models and releases from OpenAI. ['Atlas', 'ChatGPT', 'ChatGPT Enterprise', 'ChatGPT-Live', 'ChatGPT-User', 'ChatGPT Work', 'Claude Fable 5', 'CLIP', 'Codex', 'Codex CLI', 'Codex Micro', 'DALL-E', 'DALL-E 2', 'DALL-E 3', 'GPT', 'GPT-2', 'GPT2', 'GPT-3', 'GPT 3.5', 'GPT-3.5', 'gpt-3.5-turbo', 'GPT-3.5-turbo', 'GPT-3.5 Turbo', 'GPT-3-XL', 'GPT-4', 'gpt-4.1', 'GPT-4.1', 'gpt-4.1-mini', 'GPT-4 (ChatGPT)', 'gpt-4o', 'GPT-4o', 'GPT-4o mini', 'GPT-4 Turbo', 'GPT-4V(ision)', 'GPT-5', 'gpt-5.1', 'GPT-5.1', 'GPT 5.2', 'GPT-5.2', 'GPT-5.2-codex']

Official site ↗

Research 🇺🇸

DeepSeek hints at R2 model and introduces new inference scaling method SPCT

DeepSeek has published a study proposing the Self-Principled Critique Tuning (SPCT) method to improve the scalability of general reward models during inference. At the same time, the company hinted at the upcoming release of its next model, R2, which is expected to be based on pure reinforcement learning approaches.

DeepSeek

OpenAI

Synced28.07 · 00:03

Models 🇷🇺

How Kimi K3 Works: 2.8 Trillion Parameters, Linear Attention, and Million-Token Agents

Moonshot AI has released the weights of its Kimi K3 model, a multimodal model based on a Mixture of Experts architecture with 2.8 trillion parameters, of which 104 billion are activated per token. In certain programming and tool-use benchmarks, the model outperforms competitors, but on average it lags behind some of them.

Moonshot AI

Anthropic

OpenAI

Habr — хаб ИИ27.07 · 23:03

Show more ↓