Models

Works best with detailed instructions and stacked skills.

A high-quality 8-bit (Q8_0) quantization of Microsoft's Phi-1.5 (1.3B). Optimized for local coding assistance and logical reasoning with a tiny 1.8GB memory footprint. Features custom stop tokens to prevent base-model hallucinations.

ai-human-lab/EEVE-Korean_Instruct-10.8B-expo

4-bit quantized model of Uzbek open-source model on HuggingFace at uzlm/alloma-1B-Instruct

This is a fine-tuned llama 3.2 model with 1 billion parameters. It turns plain English questions into SQL queries. It was trained using examples of questions and their matching SQL, making it easier to work with databases using natural language.

KoboldAI/LLaMA2-13B-Tiefighter

Modelo de IA especializado na Doutrina Espírita de Allan Kardec. Treinado sobre Llama 3.1 8B Instruct com QLoRA, cobrindo a obra integral de Kardec em domínio público: as 5 obras da Codificação + obras complementares + 12 volumes da Revista Espírita

For language translation. A GGUF model import from https://huggingface.co/TheBloke/ALMA-13B-GGUF

EvoluMK2: AEGIS-IX is a 9-model suite (1.5B–5B) engineered by wallpillar-lm. It combines zero-fluff stoic execution, anti-jailbreak immunity, and dynamic reasoning to deliver high-density technical analysis.

tarek-abdelhak-model is a fine-tuned version of SmolLM2-1.7B-Instruct trained to answer questions about Tarek Abdelhak, a Mid-Senior Frontend Engineer with 4 years of professional experience.

Arctic-Text2SQL-R1-7B is a 7-billion-parameter Text-to-SQL model fine-tuned using Group Relative Policy Optimization (GRPO) with a simple execution-based reward signal. It converts natural language questions into executable SQL queries.

Source: https://huggingface.co/mradermacher/LFM2.5-1.2B-MEGABRAIN2-Thinking-Kimi-V2-DISTILL-GGUF

SILMA Kashif 2B Instruct v1.0 bnb 4bit Quantized

A Q4_K_M GGUF of Gemma 4 E2B, fine-tuned on Apple Silicon (M3 Pro, 18 GB) using MLX LoRA to act as an autonomous AIOps Orchestrator Agent.

Pure CPU fine-tune for day to day talk, help, documentation. FR. Approx 16 GB ram. Friendly, with sometimes a bit of fun.