LLM EXPLORER 60,702 MODELS INDEXED

Qwen2.5 2X11B CODER Dueling Wolverines V2 28B by DavidAU

By DavidAU · 9 downloads
HF Repo ↗ Copy URL

Qwen2.5 2X11B CODER Dueling Wolverines V2 28B is an open-source language model by DavidAU. Features: 28b LLM, VRAM: 56GB, Context: 32K, License: apache-2.0, MoE, Code Generating, LLM Explorer Score: 0.19.

  2x11b   Autotrain compatible Base model:merge:open-r1/olymp... Base model:merge:qwen/qwen2.5-... Base model:open-r1/olympiccode... Base model:qwen/qwen2.5-coder-...   Chat   Code   Code generation   Codegen   Codeqwen   Coder   Coding   Conversational   En   Endpoints compatible   Merge   Mixture of experts   Moe   Qwen   Qwen-coder   Qwen2   Qwen2 moe   Qwen2moe   Region:us   Safetensors   Sharded   Tensorflow   Wolverine coder
Model Card on HF 🤗: https://huggingface.co/DavidAU/Qwen2.5-2X11B-CODER-Dueling-Wolverines-V2-28B 

Qwen2.5 2X11B CODER Dueling Wolverines V2 28B Parameters and Internals

LLM NameQwen2.5 2X11B CODER Dueling Wolverines V2 28B
Repository 🤗https://huggingface.co/DavidAU/Qwen2.5-2X11B-CODER-Dueling-Wolverines-V2-28B 
Base Model(s)  Qwen/Qwen2.5-Coder-7B   OlympicCoder 7B   Qwen/Qwen2.5-Coder-7B   open-r1/OlympicCoder-7B
Model Size28b
Required VRAM56 GB
Updated2025-09-23
MaintainerDavidAU
Model Typeqwen2_moe
Model Files  4.9 GB: 1-of-12   4.9 GB: 2-of-12   5.0 GB: 3-of-12   5.0 GB: 4-of-12   5.0 GB: 5-of-12   5.0 GB: 6-of-12   5.0 GB: 7-of-12   5.0 GB: 8-of-12   5.0 GB: 9-of-12   5.0 GB: 10-of-12   4.9 GB: 11-of-12   1.3 GB: 12-of-12
Supported Languagesen
Generates CodeYes
Model ArchitectureQwen2MoeForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.52.0.dev0
Tokenizer ClassQwen2Tokenizer
Padding Token<|im_end|>
Vocabulary Size152064
Torch Data Typebfloat16
Errorsreplace