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

Dolphin 2.2 Yi 34B 5.0bpw H6 EXL2 by LoneStriker

 »  All LLMs  »  LoneStriker  »  Dolphin 2.2 Yi 34B 5.0bpw H6 EXL2   URL Share it on

  5   

Dolphin 2.2 Yi 34B 5.0bpw H6 EXL2 is an open-source language model by LoneStriker. Features: 34b LLM, VRAM: 22.2GB, Context: 16K, License: other, Quantized, Instruction-Based, Merged, LLM Explorer Score: 0.11.

  Merged Model   Autotrain compatible   Dataset:ehartford/dolphin Dataset:ehartford/samantha-dat... Dataset:ehartford/wizardlm evo... Dataset:jondurbin/airoboros-2....   En   Endpoints compatible   Exl2   Instruct   Llama   Pytorch   Quantized   Region:us   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/LoneStriker/dolphin-2.2-yi-34b-5.0bpw-h6-exl2 

Dolphin 2.2 Yi 34B 5.0bpw H6 EXL2 Benchmarks

LLME Score: 0.11083
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Dolphin 2.2 Yi 34B 5.0bpw H6 EXL2 (LoneStriker/dolphin-2.2-yi-34b-5.0bpw-h6-exl2)
🌟 Advertise your project 🚀

Dolphin 2.2 Yi 34B 5.0bpw H6 EXL2 Parameters and Internals

Model Type 
Text Generation
Use Cases 
Areas:
Research, Commercial applications
Applications:
Multi-turn conversation, Personal advice generation
Primary Use Cases:
Empathy, Long conversations
Limitations:
Lacks alignment and bias control
Considerations:
Implement an alignment layer before exposing the model as a service.
Additional Notes 
New in 2.2: conversation and empathy features added with updates from Samantha and WizardLM data. Sponsorship by a16z. Uncensored model with an advising nature.
Supported Languages 
input (English), output (English)
Training Details 
Data Sources:
ehartford/dolphin, jondurbin/airoboros-2.2.1, ehartford/samantha-data, ehartford/WizardLM_evol_instruct_V2_196k_unfiltered_merged_split
Methodology:
Modified dataset for uncensoring, deduping, cleaning, and quality. Additional datasets added for creativity and multi-turn conversation training
Training Time:
3 days
Hardware Used:
4x A100s
Input Output 
Input Format:
ChatML format
Accepted Modalities:
Text
Output Format:
Text
LLM NameDolphin 2.2 Yi 34B 5.0bpw H6 EXL2
Repository 🤗https://huggingface.co/LoneStriker/dolphin-2.2-yi-34b-5.0bpw-h6-exl2 
Merged ModelYes
Model Size34b
Required VRAM22.2 GB
Updated2025-09-23
MaintainerLoneStriker
Model Typellama
Instruction-BasedYes
Model Files  8.6 GB: 1-of-3   8.5 GB: 2-of-3   5.1 GB: 3-of-3
Supported Languagesen
Quantization Typeexl2
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length16384
Model Max Length16384
Transformers Version4.34.1
Tokenizer ClassYiTokenizer
Padding Token<unk>
Vocabulary Size64000
Torch Data Typefloat16

Best Alternatives to Dolphin 2.2 Yi 34B 5.0bpw H6 EXL2

Best Alternatives
Context / RAM
Downloads
Likes
...2.2 Yi 34B 200K 6.0bpw H6 EXL2195K / 26.5 GB61
....2 Yi 34B 200K 4.65bpw H6 EXL2195K / 20.8 GB02
CodeLlama 34B Instruct Fp1616K / 67.5 GB28196
CodeLlama 34B Instruct Hf 4bit16K / 19.4 GB682
...phin 2.2 Yi 34B 8.0bpw H8 EXL216K / 34.8 GB63
...phin 2.2 Yi 34B 4.0bpw H6 EXL216K / 18 GB13
...phin 2.2 Yi 34B 6.0bpw H6 EXL216K / 26.4 GB52
Bagel DPO 34B V0.5195K / 68.7 GB805517
Dolphin 2.2 Yi 34B 200K195K / 69.2 GB93336
Bagel 34B V0.5195K / 68.7 GB178
Note: green Score (e.g. "73.2") means that the model is better than LoneStriker/dolphin-2.2-yi-34b-5.0bpw-h6-exl2.

Rank the Dolphin 2.2 Yi 34B 5.0bpw H6 EXL2 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? 51638 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.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20241124
  Support LLM Explorer