# Conversational AI

This is a text generation model built based on the transformers library, with an efficient parameter architecture and excellent text generation capabilities.

Sailor2 is a multilingual large language model focused on the Southeast Asian region, supporting 15 languages including English, Chinese, and various Southeast Asian languages.

Second-generation reasoning model developed by SmilyAI Labs, featuring a sarcastic antagonist-style AI persona with structured reasoning output

This model is a fine-tuned version of google/gemma-2-9b-it for function calling tasks, with training data entirely manually annotated using the Russian version of the DiTy/function-calling dataset.

Gemma-3-27B is a quantized-optimized conversational large language model supporting advanced non-linear quantization techniques, delivering high-quality text generation capabilities.

Phi-4-mini is a quantized language model developed by the PyTorch team, featuring 8-bit embeddings, 8-bit dynamic activations, and a 4-bit weight linear layer (8da4w) quantization scheme, making it suitable for mobile deployment.

This model is a proof-of-concept fine-tuned version of the Orpheus 3B TTS (Text-to-Speech) model to support Bengali.

A Japanese visual language model trained based on the NVILA-Lite architecture, supporting image-text interaction in both Japanese and English

This is a quantized version based on the mlabonne/gemma-3-4b-it-abliterated model, using llama.cpp for imatrix quantization, suitable for image-text-to-text tasks.

This is a 1.5B-parameter Russian-English bilingual instruction fine-tuned language model optimized with the MLX framework, supporting 8-bit quantized inference

This is the GPTQ 8-bit quantized version of the Llama-3.3-70B-Instruct model, fine-tuned and optimized for conversational reasoning tasks.

This is a fine-tuned mT5 model for generating questions from Persian text.

The Jais series is an Arabic-English bilingual large language model, specially optimized for Arabic performance while maintaining strong English capabilities. The 590m version is one of the foundational models in this series.

nephra v1 is a model specifically designed for roleplay conversations, trained on datasets tailored for roleplay and instructional styles.

ConvAI-9b is a conversational AI model with 9 billion parameters, built by merging Zephyr-7b-beta and Mistral-7B-v0.2, suitable for interactive scenarios such as chatbots and virtual assistants.

Pinokio_v1.0 is a text generation model supporting English and Turkish, released under the Apache-2.0 license.

StarChat2 is a 16-billion-parameter programming assistant model fine-tuned based on StarCoder2, excelling in conversation and code generation tasks

OpenHercules-2.5-Mistral-7B is a fusion model based on the Mistral-7B architecture, combining the strengths of Hercules-2.5-Mistral-7B and OpenHermes-2.5-Mistral-7B, suitable for text generation and dialogue tasks.

GGUF quantized version of the Gemma 2B model, suitable for local deployment and inference

Vikhr is a bilingual (Russian and English) open-source instruction-following large language model, specifically designed for Russian-speaking users, capable of conversation and task assistance.

PersianMind is a cross-lingual Persian-English large language model developed by the University of Tehran team, excelling in multiple Persian language tasks.

Llammas is an open-source conversational model adapted for Estonian based on the Llama-2-7B model, achieving cross-lingual knowledge transfer through two-stage training

This model is a fine-tuned version of Microsoft/DialoGPT-medium, specifically trained for generating sarcastic responses.

A large-scale generative language model developed by OpenAI, capable of understanding and generating natural language text

A large language model based on Transformer architecture, supporting various natural language processing tasks

A dialogue generation model based on the DialoGPT architecture, featuring the style of anime character Asuna Yuuki