NVIDIA Nemotron 3.5 Lightning is an open 30B mixture-of-experts (MoE) model with 3B active parameters built for always-on agents.
New state of the art 70B model. Llama 3.3 70B offers similar performance compared to the Llama 3.1 405B model.
NovaForge AI – Qwen 2.5-3B Optimized A CPU-optimized, lightweight, general-purpose AI model built on Qwen 2.5-3B, designed for fast and private local inference on low-resource systems.
StarCoder2 is the next generation of transparently trained open code LLMs that comes in three sizes: 3B, 7B and 15B parameters.
LLaMA 3.1 8B Instruct model fine-tuned for advanced Wazuh security log analysis with instruction-following capabilities
LLaMA 3.1 8B Instruct model fine-tuned for advanced Wazuh security log analysis with instruction-following capabilities.
gemma 4 31B trained on conversation with strict persona and thought trace
A local medical assistant profile based on kwangsuklee/Qwen3.5-9B.Q4_K_M-Claude-4.6-Opus-Reasoning-Distilled-v2. Base model: Qwen3.5 9B, Q4_K_M quantization, described as a Claude 4.6 Opus Reasoning distilled model. This profile is additionally customize
TGAI NB — a lightweight Chinese MoE chat model family, built end-to-end by a high school student. V2 (3B): 8-expert sparse MoE, better quality for desktop. V1 (0.86B): 4-expert sparse MoE, ultra-light, smooth on low-end devices.
NovaForge AI – Llama 3.2 3B Balanced is a well-balanced, CPU-optimized AI assistant built on Meta’s latest Llama 3.2 architecture.
EXAONE Deep exhibits superior capabilities in various reasoning tasks including math and coding benchmarks, ranging from 2.4B to 32B parameters developed and released by LG AI Research.
EXAONE 3.5 is a collection of instruction-tuned bilingual (English and Korean) generative models ranging from 2.4B to 32B parameters, developed and released by LG AI Research.
Fine-tuned 3B model that deconstructs natural-language arguments into structured JSON: isolating premises, conclusions, 22 logical forms and fallacies, and suppressed premises.
LLaMA 3.1 8B Instruct model fine-tuned for AWS cloud security event analysis.
Flagship India personal-finance chat model (1.31B), live on navyam.ai. From-scratch 1.31B (~103B tokens + cooldown) then SFT on India personal-finance. Research model - not investment advice.
Euryale L3.3 70B is a model focused on creative roleplay from Sao10k. It is the successor of Euryale L3 70B v2.2.
🎨 Gepetto - Turn ideas into 3D. A Llama 3.2-based artistic Blender assistant for modeling, art direction and validated bpy scripting.
Neuris APEX is a 3B cybersecurity input-classification model fine-tuned from Ministral 3B to identify malicious, suspicious, ambiguous, and safe user-controlled inputs across common attack families.
New state of the art 70B model. Llama 3.3 70B offers similar performance compared to Llama 3.1 405B model.