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

Open Calm 7B by cyberagent

 »  All LLMs  »  cyberagent  »  Open Calm 7B   URL Share it on

  935   

Open Calm 7B is an open-source language model by cyberagent. Features: 7b LLM, VRAM: 13.9GB, Context: 2K, License: cc-by-sa-4.0, HF Score: 28.2, LLM Explorer Score: 0.14, Arc: 20.5, HellaSwag: 30.7, MMLU: 25.2, TruthfulQA: 44.2, WinoGrande: 48.5, GSM8K: 0.2.

  Dataset:cc100   Dataset:mc4   Dataset:wikipedia   Deploy:azure   Gpt neox   Ja   Japanese   Pytorch   Region:us   Sharded
Model Card on HF 🤗: https://huggingface.co/cyberagent/open-calm-7b 

Open Calm 7B Benchmarks

ARC: 20.48 vs 96.7 (so35)-78.8%
HellaSwag: 30.65 vs 95.3 (gpt4)-67.8%
MMLU: 25.22 vs 88.3 (so35)-71.4%
TruthfulQA: 44.15 vs 59 (gpt4)-25.2%
WinoGrande: 48.54 vs 87.5 (gpt4)-44.5%
GSM8K: 0.23 vs 96.4 (so35)-99.8%
LLME Score: 0.14375
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Open Calm 7B (cyberagent/open-calm-7b)
🌟 Advertise your project 🚀

Open Calm 7B Parameters and Internals

Model Type 
Transformer-based Language Model, causal-lm
Additional Notes 
This model is a fine-tuned version of OpenCALM-XX developed by CyberAgent, Inc. The model uses GPT-NeoX, a large-scale autoregressive language modeling library in PyTorch.
Supported Languages 
ja (Proficient)
Training Details 
Data Sources:
Wikipedia (ja), Common Crawl (ja)
Model Architecture:
Decoder-only
LLM NameOpen Calm 7B
Repository 🤗https://huggingface.co/cyberagent/open-calm-7b 
Model Size7b
Required VRAM13.9 GB
Updated2026-03-29
Maintainercyberagent
Model Typegpt_neox
Model Files  9.9 GB: 1-of-2   4.0 GB: 2-of-2
Supported Languagesja
Model ArchitectureGPTNeoXForCausalLM
Licensecc-by-sa-4.0
Context Length2048
Model Max Length2048
Transformers Version4.27.0.dev0
Tokenizer ClassGPTNeoXTokenizer
Padding Token<|padding|>
Vocabulary Size52224
Torch Data Typefloat16

Best Alternatives to Open Calm 7B

Best Alternatives
Context / RAM
Downloads
Likes
Literature 7B 1638416K / 36 GB915
RedPajama 7B 1638416K / 36 GB64
Stablelm Tuned Alpha 7B4K / 31.9 GB3348360
Stablelm Base Alpha 7B4K / 31.9 GB5069208
Stablelm 7B Sft V7 Epoch 34K / 32.4 GB67067
StableLManticore 7B4K / 16 GB61
Pythia 6.9B Deduped 4K4K / 27.2 GB510
Stablelm 7B4K / 31.9 GB52
RedPajama INCITE 7B Instruct2K / 13.8 GB2453108
RedPajama INCITE 7B Base2K / 13.8 GB244192
Note: green Score (e.g. "73.2") means that the model is better than cyberagent/open-calm-7b.

Rank the Open Calm 7B 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? 52721 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