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

Yi 34B 200K DARE Megamerge V8 by brucethemoose

 »  All LLMs  »  brucethemoose  »  Yi 34B 200K DARE Megamerge V8   URL Share it on

  92   

Yi 34B 200K DARE Megamerge V8 is an open-source language model by brucethemoose. Features: 34b LLM, VRAM: 68.8GB, Context: 195K, License: other, Merged, HF Score: 72.6, LLM Explorer Score: 0.12, Arc: 67.8, HellaSwag: 86.1, MMLU: 77, TruthfulQA: 56.3, WinoGrande: 82.8, GSM8K: 65.4.

  Merged Model   Arxiv:2306.01708   Arxiv:2311.03099   En   Endpoints compatible   Llama   Model-index   Region:us   Safetensors   Sharded   Tensorflow   Yi
Model Card on HF 🤗: https://huggingface.co/brucethemoose/Yi-34B-200K-DARE-megamerge-v8 

Yi 34B 200K DARE Megamerge V8 Benchmarks

ARC: 67.75 vs 96.7 (so35)-29.9%
HellaSwag: 86.06 vs 95.3 (gpt4)-9.7%
MMLU: 77.03 vs 88.3 (so35)-12.8%
TruthfulQA: 56.31 vs 59 (gpt4)-4.6%
WinoGrande: 82.79 vs 87.5 (gpt4)-5.4%
GSM8K: 65.43 vs 96.4 (so35)-32.1%
LLME Score: 0.1247
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Yi 34B 200K DARE Megamerge V8 (brucethemoose/Yi-34B-200K-DARE-megamerge-v8)
🌟 Advertise your project 🚀

Yi 34B 200K DARE Megamerge V8 Parameters and Internals

Model Type 
text-generation
Additional Notes 
Merged using DARE TIES method to handle models with up to 200,000 context efficiently. Specialized in merging multiple Yi models for improved performance.
Input Output 
Input Format:
Orca-Vicuna template
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Run at a lower temperature with 0.1 or higher MinP, a little repetition penalty, possibly mirostat with low tau.
LLM NameYi 34B 200K DARE Megamerge V8
Repository 🤗https://huggingface.co/brucethemoose/Yi-34B-200K-DARE-megamerge-v8 
Merged ModelYes
Model Size34b
Required VRAM68.8 GB
Updated2026-04-12
Maintainerbrucethemoose
Model Typellama
Model Files  9.8 GB: 1-of-7   9.8 GB: 2-of-7   9.8 GB: 3-of-7   10.0 GB: 4-of-7   9.8 GB: 5-of-7   9.8 GB: 6-of-7   9.8 GB: 7-of-7
Supported Languagesen
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length200000
Model Max Length200000
Transformers Version4.35.2
Tokenizer ClassLlamaTokenizer
Padding Token<unk>
Vocabulary Size64002
Torch Data Typebfloat16

Quantized Models of the Yi 34B 200K DARE Megamerge V8

Model
Likes
Downloads
VRAM
...4B 200K DARE Megamerge V8 GGUF142899 GB
...4B 200K DARE Megamerge V8 GPTQ3918 GB
...34B 200K DARE Megamerge V8 AWQ2519 GB

Best Alternatives to Yi 34B 200K DARE Megamerge V8

Best Alternatives
Context / RAM
Downloads
Likes
Bagel Hermes 34B Slerp195K / 68.9 GB80731
34B Beta195K / 69.2 GB821766
Smaug 34B V0.1195K / 69.2 GB817964
Yi 34B 200K195K / 68.9 GB10003320
Casual Magnum 34B195K / 68.8 GB71
Bagel 34B V0.2195K / 68.7 GB275241
Yi 34B 200K AEZAKMI V2195K / 69.2 GB53712
Smaug 34B V0.1 ExPO195K / 69.2 GB77380
Bagel DPO 34B V0.5195K / 68.7 GB805517
Faro Yi 34B195K / 69.2 GB80376
Note: green Score (e.g. "73.2") means that the model is better than brucethemoose/Yi-34B-200K-DARE-megamerge-v8.

Rank the Yi 34B 200K DARE Megamerge V8 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? 53053 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 v20260328a
  Support LLM Explorer