Al

[Al @albertchen121]

@sama AI BigBang💥，Love you💗

[Al @albertchen121]

@JeffDean A new model from china.

[Al @albertchen121]

@JeffDean This model doesn't look as strong as Kimi k1.5 yet.

[Al @albertchen121]

@demishassabis A new model from china.

[Al @albertchen121]

@demishassabis This model doesn't look as strong as Kimi k1.5 yet.

[Al @albertchen121]

@zibuyu9 @OpenBMB Thanks, it seems that your team is the leader of China's LLMs.

[Al @albertchen121]

@xiaohu This is truly Open AI,not closed AI.
@elonmusk
@sama
@OpenAI
@xai

[小互 @xiaohu]

面壁智能推出 MiniCPM-o 2.6

一个 8B 大小、匹敌GPT-4o 级别的全能模型，可在移动设备上运行

- 领先的视觉能力：超越了GPT-4o-202405、Gemini 1.5 Pro和Claude 3.5 Sonnet等

- 支持中英双语实时语音对话，并提供情感/速度/风格控制、端到端语音克隆、角色扮演等功能。

- 提供强大的 OCR 和视频理解能力

- 支持连续的视频和音频流输入，实现实时直播与多模态上下文理解。

- 多语言支持： 支持超过 30 种语言的输入和输出，涵盖文本和语音。

[Al @albertchen121]

@loongxl @OpenBMB It's your chance！

[Al @albertchen121]

@supermuscle5 @OpenBMB Wow, a very vivid metaphor!

[Al @albertchen121]

@zibuyu9 @OpenBMB What is densing law？

[Al @albertchen121]

@LindaPo88554546 @OpenBMB @elonmusk @sama @OpenAI @xai ahhh

[OpenBMB @OpenBMB]

💥 Introducing MiniCPM-o 2.6: An 8B size, GPT-4o level Omni Model runs on device
✨ Highlights:
~Match GPT-4o-202405 in vision, audio and multimodal live streaming
~End-to-end real-time bilingual audio conversation ~Voice cloning & emotion control
~Advanced OCR & video understanding
~Offline iPad-compatible multimodal live streaming
🔗 Try it out:
GitHub:https://t.co/gtRJoHOlfd
HF:https://t.co/IY9KgoOqSI
Demo:https://t.co/IzZuyz0qB1

[AIGCLINK @aigclink]

酷！清华大学等研究团队出了新项目，主动型AI助手：ProActive Agent，它可以在没有明确指令的情况下，基于环境观察主动预测和提供帮助

它可以在收到会议邀请时自动询问是否需要安排日程、编程时主动提供代码建议、写作时主动提供相关参考资料

ProActive Agent基于数据驱动构建主动性agent，建了一个包含真实人类活动的ProactiveBench数据集（6,790条训练数据）来生成任务预测，研究团队还做了一个奖励模型来评估主动性agent的表现

经过主动性训练的模型性能优于原始模型，Qwen2-7B在经过训练后F1分数由60.74%升到66.47%，LLaMA-3.1-8B由55.06%升到66.25%

#ProActiveAgent #agent #主动型agent

[OpenBMB @OpenBMB]

🚀 Excited to share our latest work: “RAGEval”!
🎉 It’s a versatile framework for generating scenario-specific RAG evaluation datasets, complete with comprehensive metrics. Perfect for rapid evaluations! 🔍
✨ Paper: https://t.co/9MIfMkwVcG

[Al @albertchen121]

人类早期驯化AI珍贵录像

[Al @albertchen121]

人类早期驯化AI珍贵录像

[Mira Murati @miramurati]

I shared the following note with the OpenAI team today.