# Chain-of-Thought Reasoning

MMaDA is a novel class of multimodal diffusion foundation models, excelling in various domains such as text reasoning, multimodal understanding, and text-to-image generation.

DeepSeek-R1-Llama-8B-F32-GGUF is the quantized version of DeepSeek-R1-Distill-Llama-8B, trained directly with reinforcement learning, featuring capabilities such as self-verification, reflection, and generating extended chain-of-thought reasoning.

This is a quantized version of the Sarvam-m model, supporting text generation tasks in multiple Indian languages and English.

QwenStoryteller is a storytelling model based on a vision-language model, focusing on visual narrative and cross-frame consistency.

Quantized version of Qwen's visual storytelling model, focusing on cross-frame consistent story generation and image-to-text tasks

Qwen3-235B-A22B is the latest generation large language model in the Qwen series, adopting a Mixture of Experts (MoE) architecture with 235 billion parameters and 22 billion active parameters. It excels in reasoning, instruction following, agent capabilities, and multilingual support.

MAI-DS-R1 is the DeepSeek-R1 inference model, further trained by Microsoft's AI team to enhance its responsiveness on restricted topics and optimize its risk performance while maintaining its reasoning capabilities and competitive performance.

VideoChat-R1_7B_caption is a multimodal video-text generation model based on Qwen2-VL-7B-Instruct, focusing on video content understanding and description generation.

UIGEN-T1.5 is an advanced Transformer UI generation model fine-tuned from Qwen2.5-Coder-32B-Instruct, specifically designed for generating modern and unique front-end user interfaces.

An advanced UI generation model fine-tuned based on Qwen2.5-Coder-7B-Instruct, excelling in generating modern and visually appealing HTML/CSS code.

An advanced Transformer-based interface generation model fine-tuned on Qwen2.5-Coder-14B-Instruct, specializing in generating modern and unique frontend user interfaces

14B-parameter-scale mathematical reasoning large language model supporting chain-of-thought reasoning

DeepHermes 3 Preview Abliterated is the latest version of Nous Research's flagship Hermes series large language model, integrating reasoning and conventional LLM response modes.

Gemma-3-4b Reasoning is a Transformer-based language model fine-tuned using the GRPO method, specializing in reasoning task optimization.

TraceBack 12b is a 4bit quantized version based on the Mistral-Nemo-Instruct architecture, focusing on instruction-following and chain-of-thought reasoning tasks.

Tifa-DeepsexV2-7b-MGRPO-safetensors is a multilingual (Chinese and English) large language model based on the transformers library, optimized through incremental pre-training, supervised fine-tuning, and reinforcement learning, suitable for role-playing and chain-of-thought tasks.

MiniThinky 1B is an AI assistant model fine-tuned from Llama-3.2-1B-Instruct, featuring thought process visualization before answering

A safety protection tool fine-tuned based on internlm2-7b-chat, providing human-readable judgment explanations and fine-grained harm scoring

A targeted sentiment analysis model fine-tuned from Flan-T5-XL, trained using the Three-Hop Reasoning (THoR) framework, specifically designed for English text

A fine-tuned model based on llama 3 8b, focusing on system prompt fidelity, helpfulness, and role immersion

Based on the Mistral-7b model, it employs the Quiet-STaR method for continuous pretraining, generating 8 reasoning tokens before each output token to enhance reasoning capabilities.

An NLI model fine-tuned based on flan-t5-xxl, used to verify whether a premise supports a hypothesis, specially optimized for multi-sentence premise scenarios

Synthia-70B-v1.5 is a 70-billion-parameter large language model based on the Llama2 architecture, focusing on complex reasoning and coherent responses through the Tree of Thought method.

Llemma 34B is a language model specialized in the field of mathematics, initialized based on the weights of Code Llama 34B and trained on 50 billion tokens from the Proof-Pile-2 dataset.

SynthIA (Synthetic Intelligence Agent) is an LLama-2-70B model trained on an Orca-style dataset, excelling in instruction following and long dialogues.