Brief Details: Azure Dusk v0.2 GGUF quantized model with multiple compression variants (3.1GB-10.2GB) optimized for different performance/size tradeoffs
Brief-details: GGUF quantized version of Mistral-Nemo-Instruct model optimized for different performance/size tradeoffs, featuring multiple compression levels from Q2 to Q8
BRIEF-DETAILS: Quantized version of INTELLECT-MATH model optimized for mathematical tasks, offering various compression levels from 2GB to 6.4GB with IQ/Q variants
Brief-details: INTELLECT-MATH-GGUF is a quantized mathematics-focused language model available in multiple GGUF formats, optimized for different performance/size trade-offs
BRIEF-DETAILS: A comprehensive guide for managing Mixture of Experts (MOE) model activation in various LLM applications, with detailed instructions for setting expert counts and implementation across different platforms.
BRIEF-DETAILS: Qwen2.5-14B quantized model offering various GGUF formats from 3.7GB to 12.2GB, optimized for different performance/size tradeoffs. Features imatrix quantization.
BRIEF DETAILS: GGUF quantized version of BuddyGlassUncensored2025.6 offering multiple compression options (Q2_K to Q8_0) with sizes ranging from 9GB to 25.2GB. Optimized for different performance/quality tradeoffs.
Brief Details: Specialized 8B parameter LLM fine-tuned from Llama-3.1 for converting natural language to Cypher queries, optimized for Neo4j database interactions.
Brief-details: A high-performance quantized version of Mistral-Small-24B offering various compression levels from 25GB to 7GB, optimized for different hardware configurations and use cases.
Brief-details: Quantized version of Viper-Coder optimized for coding tasks, offering multiple compression variants (3.7GB-12.2GB) with IQ/Q formats for different performance needs
Brief Details: A testing-focused tiny random implementation of Stable Diffusion 3, designed for internal validation and optimization purposes.
Brief-details: Qwen2.5-Coder-32B-Instruct-GPTQ-Int4 is a powerful 32B parameter code-focused LLM with 4-bit quantization, 128K context length, and state-of-the-art coding capabilities.
BRIEF DETAILS: Neural machine translation model for English to Indonesian, by Helsinki-NLP. Features transformer architecture with SentencePiece tokenization. BLEU: 38.3
Brief-details: English to Northern Sotho translation model fine-tuned from Helsinki-NLP/opus-mt-en-nso, achieving 18.93 BLEU score with improved translation quality
BRIEF DETAILS: MLX-optimized 1.5B parameter distilled version of DeepSeek-R1-Qwen model, offering efficient performance on Apple Silicon devices
Brief-details: A tiny random variant of SigLIP (Sigmoid-based Language-Image Pre-training) model, developed by katuni4ka for experimental purposes in vision-language tasks.
Brief-details: Phi-3-Mini-4K-Instruct is a 3.8B parameter lightweight LLM optimized for reasoning and instruction following, with 4K context window and state-of-the-art performance for its size.
Brief-details: EXD-v1 is an AI model by Jonny001, hosted on HuggingFace. Shows connections to NSFW content filtering and Alita model architectures, suggesting potential content moderation applications.
Brief Details: iroiroLoRA is a specialized LoRA (Low-Rank Adaptation) model created by nashikone, available on Hugging Face for fine-tuning applications.
BRIEF-DETAILS: Powerful 10.7B parameter LLM optimized for chat/instruction, trained on 1M GPT-4 entries. Strong benchmark performance with GGUF quantization options from 2-8 bits.
BRIEF DETAILS: Zamba2-7B is a powerful 7B parameter hybrid model combining state-space (Mamba) and transformer architectures, offering state-of-the-art performance with efficient inference and lower memory footprint.
