Swallow 7B Instruct V0.1 by tokyotech-llm

 ยป  All LLMs  ยป  tokyotech-llm  ยป  Swallow 7B Instruct V0.1   URL Share it on

  Autotrain compatible   Conversational   En   Endpoints compatible   Instruct   Ja   Llama   Region:us   Safetensors   Sharded   Tensorflow

Swallow 7B Instruct V0.1 Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
๐ŸŒŸ Advertise your project ๐Ÿš€

Swallow 7B Instruct V0.1 Parameters and Internals

Model Type 
text-generation, instruction-tuned
Use Cases 
Areas:
Research, Development
Applications:
Cross-Lingual Adaptation, Instruction Following
Primary Use Cases:
Text Generation, Language Translation
Limitations:
Not fine-tuned for specific human intent and safety considerations
Additional Notes 
Developed by multiple team members from TokyoTech-LLM, with acknowledgements to Meta Research for Llama 2.
Supported Languages 
Japanese (Proficient), English (Proficient)
Training Details 
Data Sources:
OpenAssistant Conversations Dataset EN top-1 thread, OpenAssistant Conversations Dataset
Methodology:
Supervised fine-tuning (SFT)
Model Architecture:
Please refer to LLaMA-2 technical report for details on the model architecture.
Input Output 
Input Format:
~~[INST] <> {SYSTEM_PROMPT} <> {USER_MESSAGE} [/INST]
Accepted Modalities:
text
Output Format:
Strings
Performance Tips:
Adhere strictly to instruction format to maintain performance.
Release Notes 
Version:
0.1
Date:
April 26, 2024
Notes:
Release of enhanced instruction-tuned models as preview versions.
Version:
7b-plus
Date:
March 2, 2024
Notes:
Trained with approximately twice as many Japanese tokens.
Version:
13b-NVE-hf
Date:
February 4, 2024
Notes:
Model release with no vocabulary expansion.
Version:
7b-NVE
Date:
January 26, 2024
Notes:
Release of various instruct-hf models as well as no vocabulary expansion models.
Version:
7b
Date:
December 19, 2023
Notes:
Initial release of Swallow 7b, 13b, and 70b in instruct hf variants.
LLM NameSwallow 7B Instruct V0.1
Repository ๐Ÿค—https://huggingface.co/tokyotech-llm/Swallow-7b-instruct-v0.1 
Model Size7b
Required VRAM13.7 GB
Updated2025-06-09
Maintainertokyotech-llm
Model Typellama
Instruction-BasedYes
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   3.8 GB: 3-of-3
Supported Languagesen ja
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length4096
Model Max Length4096
Transformers Version4.39.0.dev0
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size43176
Torch Data Typebfloat16
Swallow 7B Instruct V0.1 (tokyotech-llm/Swallow-7b-instruct-v0.1)

Quantized Models of the Swallow 7B Instruct V0.1

Model
Likes
Downloads
VRAM
Swallow 7B Instruct V0.1 4bit1134 GB

Best Alternatives to Swallow 7B Instruct V0.1

Best Alternatives
Context / RAM
Downloads
Likes
A3.41024K / 16.1 GB130
1241024K / 16.1 GB930
A5.41024K / 16.1 GB120
A2.41024K / 16.1 GB120
... Qwen2.5llamaify 7B V23.1 200K195K / 15.2 GB15984
SuperNeuralDreadDevil 8B128K / 16.1 GB701
Falcon3 7B Instruct32K / 14.8 GB4379071
Falcon3 Jessi V0.4 7B Slerp32K / 14.9 GB109
Jessi V0.4 Falcon3 7B Instruct32K / 14.8 GB60
Jessi V0.6 Falcon3 7B Instruct32K / 14.8 GB90
Note: green Score (e.g. "73.2") means that the model is better than tokyotech-llm/Swallow-7b-instruct-v0.1.

Rank the Swallow 7B Instruct V0.1 Capabilities

๐Ÿ†˜ 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! ๐ŸŒŸ

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? 48023 in total.

Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124