Upload README.md

README.md

@@ -1,15 +1,53 @@
 ---
-title: ZeppFusion
-emoji: 💬
-colorFrom: yellow
+title: Chat with Llama 4, Gemini 2.5, Qwen2.5, Mistral and Deepseek-R1 (reasoning) augmented LLMs
+emoji: 🚀
+colorFrom: indigo
 colorTo: purple
-sdk: gradio
-sdk_version: 5.42.0
-app_file: app.py
-pinned: false
-hf_oauth: true
-hf_oauth_scopes:
-- inference-api
+sdk: docker
+pinned: true
+license: apache-2.0
+models:
+- deepseek-ai/DeepSeek-R1
+- deepseek-ai/DeepSeek-R1-Distill-Llama-70B
+- mistralai/Mistral-Small-24B-Instruct-2501
+- mistralai/Mistral-Large-Instruct-2411
+- mistralai/Mistral-Nemo-Instruct-2407
+- meta-llama/Llama-4-Scout-17B-16E-Instruct
+short_description: Supercharge your favourite LLMs with reasoning capabilities
 ---
+# Reasoner4All
 
-An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).
+Reasoner4All is a simple demo that allows you to supercharge your favourite LLMs by integrating advanced reasoning capabilities from the latest open-source thinking models. Whether you're working with OpenAI, Mistral, Groq, or Gemini, this framework seamlessly enhances their decision-making with an additional reasoning layer. Leverage models like DeepSeek R1 Distill LLaMA-70B to improve your AI's problem-solving skills and contextual understanding!
+
+## Useful Links 🔗
+
+- AiCore [Repo](https://github.com/BrunoV21/AiCore) my custom wrapper around multiple providers which uniforms streaming, json outputs and reasoning integration.
+
+## Features
+
+**Supports Multiple LLM Providers**
+- OpenAI
+- Mistral
+- Groq (including DeepSeek R1 Distill LLaMA-70B for reasoning)
+- Gemini
+- and more!
+
+**Reasoner Integration** 
+- Uses an additional (reasoning) LLM to enhance reasoning capabilities
+- Support for Deepseek R1 Distill LLaMA-70B (Groq Hosted) and Deepseek R1 (Nvidia Hosted)
+
+**Conversation History**
+- For context reasons the converstation history up to latest 4096 tokens is preserved and passed to the llms as context
+
+## User Interface
+**Settings**
+- Provider and Model selection
+- Reasoner Option and Model selection
+- Support for System Propmpt and Reasoner System prompt specification
+
+**Profiles**
+- Reasoner4All uses a series of pre-set API Keys for demo purposes
+- OpenAi allows you to connect to your models with support for api key if required *don't worry I am not logging your keys anwyehre but you can check the src code of this space to be sure :)*
+
+**Chat**
+- Reasoning steps appear inside a step which can be expanded to visualize the process