LLM EXPLORER 62,076 MODELS INDEXED

Meta Llama 3 225B Instruct by mlabonne

By mlabonne · 274 downloads
HF Repo ↗ Copy URL

Meta Llama 3 225B Instruct is an open-source language model by mlabonne. Features: 120b LLM, VRAM: 443.2GB, Context: 8K, License: other, Instruction-Based, Merged, LLM Explorer Score: 0.12.

  Merged Model Base model:finetune:mlabonne/m... Base model:mlabonne/meta-llama...   Conversational   Endpoints compatible   Instruct   Llama   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/mlabonne/Meta-Llama-3-225B-Instruct 

Meta Llama 3 225B Instruct Parameters and Internals

Additional Notes 
This model currently has stability issues, and it is not recommended for use without further testing and validation by users.
LLM NameMeta Llama 3 225B Instruct
Repository 🤗https://huggingface.co/mlabonne/Meta-Llama-3-225B-Instruct 
Base Model(s)  Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   Meta Llama 3 120B Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct   mlabonne/Meta-Llama-3-120B-Instruct
Merged ModelYes
Model Size120b
Required VRAM443.2 GB
Updated2026-07-14
Maintainermlabonne
Model Typellama
Instruction-BasedYes
Model Files  10.0 GB: 1-of-46   9.8 GB: 2-of-46   9.8 GB: 3-of-46   9.8 GB: 4-of-46   9.8 GB: 5-of-46   9.8 GB: 6-of-46   10.0 GB: 7-of-46   10.0 GB: 8-of-46   9.8 GB: 9-of-46   9.8 GB: 10-of-46   9.8 GB: 11-of-46   9.8 GB: 12-of-46   9.8 GB: 13-of-46   9.8 GB: 14-of-46   10.0 GB: 15-of-46   10.0 GB: 16-of-46   9.6 GB: 17-of-46   9.8 GB: 18-of-46   10.0 GB: 19-of-46   10.0 GB: 20-of-46   9.6 GB: 21-of-46   9.8 GB: 22-of-46   10.0 GB: 23-of-46   9.8 GB: 24-of-46   9.8 GB: 25-of-46   9.8 GB: 26-of-46   10.0 GB: 27-of-46   9.8 GB: 28-of-46   9.8 GB: 29-of-46   9.8 GB: 30-of-46   10.0 GB: 31-of-46   9.8 GB: 32-of-46   9.8 GB: 33-of-46   9.8 GB: 34-of-46   10.0 GB: 35-of-46   9.8 GB: 36-of-46   9.8 GB: 37-of-46   9.8 GB: 38-of-46   10.0 GB: 39-of-46   9.8 GB: 40-of-46   9.8 GB: 41-of-46   9.8 GB: 42-of-46   10.0 GB: 43-of-46   9.8 GB: 44-of-46   9.8 GB: 45-of-46
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length8192
Model Max Length8192
Transformers Version4.40.1
Tokenizer ClassPreTrainedTokenizerFast
Vocabulary Size128256
Torch Data Typefloat16

Best Alternatives to Meta Llama 3 225B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
...ma 3 Instruct 120B Cat A Llama8K / 243.9 GB21
...0B Instruct Abliterated Merged8K / 243.7 GB11
Koishi 120B Qlora4K / 216.9 GB71
...t 120B Cat A Llama EXL2 5.5bpw8K / 85.3 GB50
...t 120B Cat A Llama EXL2 4.5bpw8K / 70.3 GB41
...egaDolphin 120B 2.9bpw H6 EXL24K / 44.3 GB23
...gaDolphin 120B 2.65bpw H6 EXL24K / 40.5 GB32
...egaDolphin 120B 4.0bpw H6 EXL24K / 60.8 GB41
MegaDolphin 120B GPTQ4K / 61.1 GB254
MegaDolphin 120B AWQ4K / 63.3 GB72
Note: green Score (e.g. "73.2") means that the model is better than mlabonne/Meta-Llama-3-225B-Instruct.