Tess-v2.5 (Qwen2-72B) was fine-tuned over the newly released Qwen2-72B base, using the Tess-v2.5 dataset that contain 300K samples spanning multiple topics.
The latest in the Smaug series - a finetune of Qwen2-72B-Instruct
This is a new kind of model optimization. This model is based on MaziyarPanahi/calme-2.1-qwen2-72b, which was tuned from Qwen2-72B.
Security-posture cluster model: THOX configuration (not training) of Qwen2.5-72B-Instruct with a security-focused system prompt, for air-gapped and isolated deployments. Qwen Research license.
Department-scale cluster model: THOX configuration (not training) of Qwen2.5-72B-Instruct for local inference and MeshStack deployments. Qwen Research license.
A 72B parameter coding model optimized for software engineering tasks, based on the Qwen2.5-72B architecture.
AbacusAI Dracarys2 72B
This model is based on Qwen2-72b, Dolphin-2.9.2 has a variety of instruction, conversational, and coding skills. It also has initial agentic abilities and supports function calling. Dolphin is uncensored.
[GGUF] Qwen2.5 instruction-tuned language model ranging from 0.5 to 72 billion parameters.
gte-Qwen2-7B-instruct is the model in the gte (General Text Embedding).
Japanese instruction-tuned LLM by CyberAgent, distilled from Qwen-72B.
Qwen 2.5 Coder 7b Finetuned on MainframeBench for Cobol Coding
Text + Vision Qwen3.8-27B-FableColdFusion-735882-HereticUncensored-NEOCODERMAX-MTP-NVFP4-Q8-Q3
This is a simple abliterated (refusal-orthoganalized) version of the Qwen2-7B-Instruct model.
Qwen2-7B-Instruct with OpenDevin Tool Calling
Qwen2.5VL-7B-Instruct-Q5_K_M is a vision-language model from Alibaba Cloud with 7 billion parameters, designed for processing text and visual inputs, and optimized with Q5_K_M quantization for efficient deployment in ollama.
unsloth/Qwen2.5-7B-Instruct-1M
Qwen2.5-Coder-7B fine-tuned for uncensored code generation. Made by RACER IS OP.