LLM EXPLORER 59,420 MODELS INDEXED

Calm3 22B Chat by cyberagent

By cyberagent · 878 downloads
HF Repo ↗ Copy URL

Calm3 22B Chat is an open-source language model by cyberagent. Features: 22b LLM, VRAM: 44.9GB, Context: 16K, License: apache-2.0, LLM Explorer Score: 0.21.

  Conversational   Deploy:azure   En   Ja   Japanese   Llama   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/cyberagent/calm3-22b-chat 

Calm3 22B Chat Parameters and Internals

Model Type 
Transformer-based Language Model
Supported Languages 
Japanese (N/A), English (N/A)
Training Details 
Data Volume:
2.0 trillion tokens
Context Length:
16384
Model Architecture:
Decoder-only
Input Output 
Input Format:
CALM3-Chat uses ChatML as the prompt format.
LLM NameCalm3 22B Chat
Repository 🤗https://huggingface.co/cyberagent/calm3-22b-chat 
Model Size22b
Required VRAM44.9 GB
Updated2026-08-02
Maintainercyberagent
Model Typellama
Model Files  4.9 GB: 1-of-10   4.9 GB: 2-of-10   4.8 GB: 3-of-10   4.9 GB: 4-of-10   5.0 GB: 5-of-10   4.8 GB: 6-of-10   4.9 GB: 7-of-10   4.8 GB: 8-of-10   4.9 GB: 9-of-10   1.0 GB: 10-of-10
Supported Languagesja en
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length16384
Model Max Length16384
Transformers Version4.42.3
Tokenizer ClassGPTNeoXTokenizer
Padding Token<|padding|>
Vocabulary Size65024
Torch Data Typebfloat16

Best Alternatives to Calm3 22B Chat

Best Alternatives
Context / RAM
Downloads
Likes
EuroLLM 22B Instruct 251232K / 45.5 GB334263
EuroLLM 22B 251232K / 45.5 GB276619
Calm3 22B RP V216K / 44.9 GB2120
Calm3 22B RP V0.116K / 44.9 GB100
Yousei 22B4K / 44.5 GB972
Llama2 22B Daydreamer V34K / 43.7 GB17711
Platypus 2 22B Relora4K / 43.7 GB931
Llama2 22B Blocktriangular4K / 43.7 GB3684
Llama2 22B4K / 43.7 GB10946
Llama2 22B Daydreamer V24K / 43.7 GB72
Note: green Score (e.g. "73.2") means that the model is better than cyberagent/calm3-22b-chat.