# High-precision Quantization

A quantized version based on the cognitivecomputations/Qwen3-72B-Embiggened model, quantized using llama.cpp, and can run efficiently in various environments.

A Llamacpp imatrix quantized version based on allura-org/Q3-30B-A3B-Designant, suitable for various quantization needs, supporting role-playing and conversational tasks.

A 12B-parameter multilingual large language model based on llama.cpp quantization, supporting role-play, story creation, and multi-domain professional tasks

Q3-30b-A3b-Pentiment is a large language model based on the LLaMA architecture, optimized through quantization for various text generation tasks.

Quantized version of INTELLECT-2, optimized using llama.cpp, supporting multiple quantization types to accommodate different hardware requirements.

Llamacpp imatrix quantized version of Dolphin-Mistral-24B-Venice-Edition, supporting multiple quantization types, suitable for text generation tasks.

This is a quantized version of the Qwen3-8B model, using llama.cpp for iMatrix quantization, suitable for chat scenarios.

Remnant-GLM4-32B is a 32B-parameter large language model based on the GLM4 architecture, supporting role-playing and conversational interactions, particularly suitable for salamander-related applications.

Quantized version based on Qwen/Qwen3-30B-A3B, using llama.cpp for multi-precision quantization, suitable for text generation tasks.

A 9B-parameter chat model based on GLM-4 architecture, supporting Chinese and English dialogues, quantized for various hardware environments

A large language model based on 12B parameters, supporting text generation tasks, released under the Apache-2.0 license.

This is a 24B-parameter large language model, processed with llama.cpp's imatrix quantization, offering multiple quantized versions to suit different hardware requirements.

A dialogue model fine-tuned based on Llama-3-8B, supporting multiple quantization versions, suitable for tasks such as chatting, reasoning, and role-playing.

A quantized version based on NVIDIA's AceInstruct-7B model, processed using llama.cpp with support for multiple quantization types, suitable for tasks in code, mathematics, and general domains.

Dolphin3.0-R1-Mistral-24B is a 24B-parameter large language model based on the Mistral architecture, trained by Eric Hartford, focusing on reasoning and first-principles analysis.

GGUF quantized version of DeepSeek-R1-Distill-Llama-70B-abliterated, suitable for local inference, offering multiple quantization options to meet different hardware requirements.

DeepSeek-R1-Distill-Qwen-32B-abliterated is a distilled version based on Qwen-32B, offering multiple quantization options to accommodate different hardware requirements.