LLM EXPLORER 67,952 MODELS INDEXED

Wizard Mega 13B by openaccess-ai-collective

By openaccess-ai-collective · 127 downloads

Wizard Mega 13B is an open-source language model by openaccess-ai-collective. Features: 13b LLM, VRAM: 26GB, Context: 2K, Instruction-Based, LLM Explorer Score: 0.09, Arc: 57.3, HellaSwag: 81.1, MMLU: 50.6, GSM8K: 10.1.

Dataset:anon8231489123/sharegp... Dataset:ehartford/wizard vicun... Dataset:ehartford/wizardlm alp...   En   Endpoints compatible   Instruct   Llama   Pytorch   Region:us   Sharded

Wizard Mega 13B Parameters and Internals

Model Type 
text-generation
Additional Notes 
Model can produce problematic outputs. Not aligned to human preferences or using RLHF.
Supported Languages 
en (Proficient)
Training Details 
Data Sources:
ShareGPT, WizardLM, Wizard-Vicuna
Methodology:
Fine-tuning
Training Time:
15 hours
Hardware Used:
8xA100 80GB
Release Notes 
Version:
Epoch Two
Notes:
Model epoch two chosen as final release due to eval loss increase during third epoch.
LLM NameWizard Mega 13B
Repository πŸ€—https://huggingface.co/openaccess-ai-collective/wizard-mega-13b 
Model Size13b
Required VRAM26 GB
Updated2026-08-08
Maintaineropenaccess-ai-collective
Model Typellama
Instruction-BasedYes
Model Files  9.9 GB: 1-of-3   9.9 GB: 2-of-3   6.2 GB: 3-of-3
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Context Length2048
Model Max Length2048
Transformers Version4.30.0.dev0
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size32000
Torch Data Typebfloat16

Quantized Models of the Wizard Mega 13B

Model
Likes
Downloads
VRAM
Wizard Mega 13B GPTQ108547 GB
Wizard Mega 13B GGUF44135 GB
Wizard Mega 13B AWQ3117 GB

Best Alternatives to Wizard Mega 13B

Best Alternatives
Context / RAM
Downloads
Likes
NexusRaven V2 13B16K / 26 GB105472
CodeLlama 13B Instruct Hf16K / 26 GB2469160
CodeLlama 13B MORepair16K / 26 GB92
CodeLlama 13B Instruct Hf16K / 26 GB36929
TableLLM 13B16K / 26 GB9834
NexusRaven 13B16K / 26 GB37106
Panda Coder 13B16K / 26 GB1114
... Llama 2 13B Instruct Text2sql16K / 26 GB828
Gen Sim16K / 0.3 GB43
Llama 3 13B Instruct Ft8K / 26.1 GB102
Note: green Score (e.g. "73.2") means that the model is better than openaccess-ai-collective/wizard-mega-13b.