Yi-34B-Chat Fine-tuned Model and Quantized Version from Lingyi Wanwu

On November 24th, ZeroOneAI officially released and open-sourced the fine-tuned model Yi-34B-Chat, which is available for free commercial use upon application. ZeroOneAI also provides developers with 4bit/8bit quantized versions of the model, where the Yi-34B-Chat4bit quantized model can be directly used on consumer-grade graphics cards. In various evaluations, Yi-34B-Chat has demonstrated outstanding performance in English knowledge level evaluation sets, Chinese comprehensive exam evaluation sets, and mathematical reasoning ability evaluation sets. The Yi series of models have been open-sourced in the ModelScope community, where developers can download, infer, and train these models.

NVIDIA Releases Its First Open-Source Full-Duplex Speech Model, VoiceChat 11B

NVIDIA has released its first open-source end-to-end full-duplex speech dialogue model, NemotronLabs VoiceChat 11B, which uses a single network to directly perform streaming speech understanding and generation, replacing the traditional ASR, LLM, TTS step-by-step architecture, achieving breakthroughs in real-time speech interaction.

Qwen3.8-Max Launch: Focused on Real Office and Complex Tasks, Model Weights to Be Open-Sourced Next Week

Alibaba released the new generation large model Qwen3.8-Max, which has been comprehensively upgraded in programming, office work, long-term tasks, and multi-modal agents. It can autonomously execute complex tasks end-to-end with minimal human intervention, completing from goal understanding to result delivery. The model weights will be open-sourced next week, focusing on enhancing the execution and delivery capabilities of agents in real scenarios.

DeepSeek V4 Official Version Tentatively Scheduled for August 3rd: Silicon Flow Price Hike Exposed, API Can Now Answer New Questions

DeepSeek V4's official release has been delayed since mid-July grayscale testing, with no date announced. SiliconFlow platform will raise the DeepSeek V4 Pro cache hit price from 0.1 to 1.0 yuan per million tokens starting Aug 3, widely seen as a signal of the imminent V4 launch, making Aug 3 a potential release date.....

Tencent WorkBuddy's Monthly Visits Exceed 20 Million in June, Leading the AI Office Intelligent Agent Market

According to the "Q2 2026 China Office Intelligent Agent Market Insight Report", Tencent WorkBuddy leads the PC-based AI-native office intelligent agent market. Its monthly visits reached 20.97 million in June, surpassing the combined total of the second and third places. The total market visits exceeded 60 million in the same month, indicating a continued expansion of the industry scale. CodeBuddy, Tencent's AI programming intelligent agent, also ranks among the top. WorkBuddy can complete office tasks through natural language commands.

Alibaba's New Generation Large Model Qwen 3.8 Is Coming: Preview Version Now Available on Alibaba Cloud and Qoder, Official Version to Be Open-Sourced Soon

China's LLM open-source pace quickens. Alibaba will soon release and open-source Qwen 3.8; its preview, Qwen3.8-Max, is live on Alibaba Cloud's Token Plan, Qoder, and QoderWork, letting users test its capabilities early. The official version follows, boosting Qwen family's open-source advancement.....