# Role-playing enhancement

An inference model based on Mistral, optimizing inference and output generation capabilities through NEO Imatrix quantization and MAX output tensors

An optimized version based on the Qwen3-30B-A3B Mixture of Experts model, improving speed by reducing the number of active experts, supporting 64k context length, and suitable for various text generation tasks.

This is a large language model merged from multiple 12B-parameter models, focusing on text generation and reasoning tasks.

This is a test model fine-tuned based on PocketDoc/Dans-SakuraKaze-V1.0.0-12b, designed to imbue existing models with a QwQ mindset, suitable for role-playing/adventure/collaborative writing scenarios.

An 8B-parameter large language model fused via MergeKit from multiple high-quality models, optimized for reasoning, dialogue, and creative generation

A 70B-parameter large language model based on the Llama 3.3 architecture, enhanced through the merging of multiple superior model components to achieve improved story narration, scene description, and prose detailing capabilities

A large language model fine-tuned based on Qwen2.5-72B, specializing in text generation and instruction-following tasks

This model is a fine-tuned version of arcee-ai/Llama-3.1-SuperNova-Lite on the CharacterAI-logs and OpenCAI-ShareGPT datasets, supporting long-context processing.

This is a series of models fine-tuned based on Qwen2.5-72B-Instruct, aiming to imitate the prose quality of the Claude 3 models (especially Sonnet and Opus).

An experimental fine-tuned model based on yi-34b-200k, suitable for creative writing, role-playing, and other tasks, without DPO stage applied.

An instruction-following model fine-tuned based on OpenLlama-3B, optimized for role-playing, instruction following, and code generation