LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

MegaDolphin 120B by cognitivecomputations

 »  All LLMs  »  cognitivecomputations  »  MegaDolphin 120B   URL Share it on

  52   
  Merged Model   Autotrain compatible Base model:cognitivecomputatio... Base model:finetune:cognitivec...   Conversational   Dataset:ehartford/dolphin Dataset:ehartford/samantha-dat... Dataset:ehartford/wizardlm evo... Dataset:jondurbin/airoboros-2....   En   Endpoints compatible   Instruct   Llama   Model-index   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/cognitivecomputations/MegaDolphin-120b 

MegaDolphin 120B Benchmarks

ARC: 69.03 vs 96.7 (so35)-28.6%
HellaSwag: 87.8 vs 95.3 (gpt4)-7.9%
MMLU: 69.26 vs 88.3 (so35)-21.6%
TruthfulQA: 59.28 vs 59 (gpt4)0.5%
WinoGrande: 81.85 vs 87.5 (gpt4)-6.5%
GSM8K: 46.25 vs 96.4 (so35)-52%
LLME Score: 0.16447
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
MegaDolphin 120B (cognitivecomputations/MegaDolphin-120b)

MegaDolphin 120B Parameters and Internals

LLM NameMegaDolphin 120B
Repository 🤗https://huggingface.co/cognitivecomputations/MegaDolphin-120b 
Base Model(s)  cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b   cognitivecomputations/dolphin-2.2-70b
Merged ModelYes
Model Size70b
Required VRAM240.7 GB
Updated2025-05-21
Maintainercognitivecomputations
Model Typellama
Instruction-BasedYes
Model Files  9.8 GB: 1-of-25   9.9 GB: 2-of-25   10.0 GB: 3-of-25   9.8 GB: 4-of-25   9.8 GB: 5-of-25   9.8 GB: 6-of-25   9.8 GB: 7-of-25   10.0 GB: 8-of-25   10.0 GB: 9-of-25   9.8 GB: 10-of-25   9.8 GB: 11-of-25   9.8 GB: 12-of-25   9.8 GB: 13-of-25   9.8 GB: 14-of-25   9.8 GB: 15-of-25   10.0 GB: 16-of-25   10.0 GB: 17-of-25   9.8 GB: 18-of-25   9.8 GB: 19-of-25   9.8 GB: 20-of-25   9.8 GB: 21-of-25   9.8 GB: 22-of-25   9.8 GB: 23-of-25   10.0 GB: 24-of-25   4.2 GB: 25-of-25
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length4096
Model Max Length4096
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Padding Token</s>
Vocabulary Size32002
Torch Data Typefloat16

Quantized Models of the MegaDolphin 120B

Model
Likes
Downloads
VRAM
MegaDolphin 120B GGUF2300 GB
MegaDolphin 120B GPTQ41161 GB
MegaDolphin 120B AWQ21763 GB

Best Alternatives to MegaDolphin 120B

Best Alternatives
Context / RAM
Downloads
Likes
... Chat 1048K Chinese Llama3 70B1024K / 141.9 GB40355
... 3 70B Instruct Gradient 1048K1024K / 141.9 GB19121
... Chat 1048K Chinese Llama3 70B1024K / 141.9 GB2655
Llama3 Function Calling 1048K1024K / 141.9 GB61
...a 3 70B Instruct Gradient 524K512K / 141.9 GB823
...a 3 70B Instruct Gradient 262K256K / 141.9 GB1355
...ama 3 70B Arimas Story RP V2.0256K / 141.1 GB103
...ama 3 70B Arimas Story RP V1.6256K / 141.2 GB90
...ama 3 70B Arimas Story RP V1.5256K / 141.2 GB122
...a 3.1 Nemotron 70B Instruct HF128K / 141.9 GB166002038
Note: green Score (e.g. "73.2") means that the model is better than cognitivecomputations/MegaDolphin-120b.

Rank the MegaDolphin 120B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 47473 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241227
  Support LLM Explorer