EmbeddingGemma 2: an open, lightweight multimodal embedding model
deepmind.google·6d ago
AI Engineering
Hacker News·Aug 26·5 min read
Qwen 3.8 27B Uncensored – Testing the Uncensored Qwen Model
TL;DR
Qwen 3.8 27B Uncensored is a hosted chat API that allows applications to handle long-context requests without needing separate model hosting. It supports both quick chats and detailed analyses through a flexible thinking mode.
✦ Why It Matters
Integrate long-context capabilities into your applications today without managing separate LLM services.
Key Takeaways
- Utilize Qwen 3.8 for long-context assistance in your applications.
- Enable thinking mode for deeper analysis of complex requests.
- Control response parameters through the imageat Chat Completions API.
- Implement moderation and logging for user-facing products.
How It Works
Qwen 3.8 operates as a hosted chat API, allowing applications to send structured messages without local model management. It processes requests by executing them on an inference server, returning responses in a format compatible with OpenAI's chat completion structure.
Users can toggle between quick responses and more thoughtful outputs by enabling or disabling the 'thinking mode', which influences the depth of the generated content.