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

Rwkv Raven 14B by RWKV

 »  All LLMs  »  RWKV  »  Rwkv Raven 14B   URL Share it on

  1020   
  Autotrain compatible   Dataset:eleutherai/pile   Endpoints compatible   Pytorch   Region:us   Rwkv   Sharded
Model Card on HF 🤗: https://huggingface.co/RWKV/rwkv-raven-14b 

Rwkv Raven 14B Benchmarks

MMLU Pro: 1.67
MUSR: 7.19
BBH: 6.76
IFEval: 7.68 vs 88 (so35)-91.3%
ARC: 44.62 vs 96.7 (so35)-53.9%
HellaSwag: 71.25 vs 95.3 (gpt4)-25.2%
MMLU: 25.92 vs 88.3 (so35)-70.6%
TruthfulQA: 41.93 vs 59 (gpt4)-28.9%
WinoGrande: 66.69 vs 87.5 (gpt4)-23.8%
GSM8K: 2.12 vs 96.4 (so35)-97.8%
MATH Lvl 5: 0.45
LLME Score: 0.14728
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Rwkv Raven 14B (RWKV/rwkv-raven-14b)
🌟 Advertise your project 🚀

Rwkv Raven 14B Parameters and Internals

Model Type 
RNN, Transformer
Additional Notes 
RWKV is an RNN with transformer-level LLM performance. It is designed for fast inference, VRAM efficiency, and 'infinite' context length. The model allows for free sentence embedding and fast training.
Input Output 
Input Format:
text
Accepted Modalities:
text
Output Format:
text
Performance Tips:
The 'Raven' models need to be prompted in a specific way. Learn more about that in the integration blog post.
LLM NameRwkv Raven 14B
Repository 🤗https://huggingface.co/RWKV/rwkv-raven-14b 
Model Size14b
Required VRAM56.9 GB
Updated2025-07-25
MaintainerRWKV
Model Typerwkv
Model Files  2.0 GB: 1-of-30   2.0 GB: 2-of-30   2.0 GB: 3-of-30   2.0 GB: 4-of-30   1.7 GB: 5-of-30   1.9 GB: 6-of-30   2.0 GB: 7-of-30   2.0 GB: 8-of-30   2.0 GB: 9-of-30   1.7 GB: 10-of-30   1.9 GB: 11-of-30   2.0 GB: 12-of-30   2.0 GB: 13-of-30   2.0 GB: 14-of-30   1.7 GB: 15-of-30   1.9 GB: 16-of-30   2.0 GB: 17-of-30   2.0 GB: 18-of-30   2.0 GB: 19-of-30   1.7 GB: 20-of-30   1.9 GB: 21-of-30   2.0 GB: 22-of-30   2.0 GB: 23-of-30   2.0 GB: 24-of-30   1.7 GB: 25-of-30   1.9 GB: 26-of-30   2.0 GB: 27-of-30   2.0 GB: 28-of-30   1.9 GB: 29-of-30   1.0 GB: 30-of-30
Model ArchitectureRwkvForCausalLM
Transformers Version4.29.0.dev0
Tokenizer ClassGPTNeoXTokenizer
Vocabulary Size50277
Torch Data Typefloat32

Best Alternatives to Rwkv Raven 14B

Best Alternatives
Context / RAM
Downloads
Likes
Rwkv 4 14B Pile0K / 56.9 GB10303
Rwkv 4 Raven 14B0K / 56.9 GB72
Rwkv 14B Wizardlm0K / 28.3 GB59

Rank the Rwkv Raven 14B 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? 50035 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 v20241124
  Support LLM Explorer