LLM EXPLORER 61,491 MODELS INDEXED

Swallow 7B Instruct GGUF by TheBloke

By TheBloke · 600 downloads
HF Repo ↗ Copy URL

Swallow 7B Instruct GGUF is an open-source language model by TheBloke. Features: 7b LLM, VRAM: 2.9GB, License: llama2, Quantized, Instruction-Based, LLM Explorer Score: 0.11.

Base model:quantized:tokyotech... Base model:tokyotech-llm/swall...   En   Gguf   Instruct   Ja   Llama   Quantized   Region:us
Model Card on HF 🤗: https://huggingface.co/TheBloke/Swallow-7B-Instruct-GGUF 

Swallow 7B Instruct GGUF Parameters and Internals

Model Type 
text-generation
Use Cases 
Limitations:
Models are still in the early stages and have not been tuned to ensure outputs align with human intent and safety considerations.
Additional Notes 
The Swallow model includes continuous pre-training from the Llama 2 family and is especially optimized for Japanese language tasks.
Supported Languages 
en (English), ja (Japanese)
Training Details 
Data Sources:
Japanese Wikipedia, RefinedWeb, Swallow Corpus, The Pile
Methodology:
The tuned versions use supervised fine-tuning (SFT)
Model Architecture:
Refer to LLaMA-2 technical report for details on the model architecture.
Responsible Ai Considerations 
Mitigation Strategies:
These models have not been tuned to ensure outputs align with human intent and safety considerations.
LLM NameSwallow 7B Instruct GGUF
Repository 🤗https://huggingface.co/TheBloke/Swallow-7B-Instruct-GGUF 
Model NameSwallow 7B Instruct
Model Creatortokyotech-llm
Base Model(s)  Swallow 7B Instruct Hf   tokyotech-llm/Swallow-7b-instruct-hf
Model Size7b
Required VRAM2.9 GB
Updated2026-08-02
MaintainerTheBloke
Model Typellama
Instruction-BasedYes
Model Files  2.9 GB   3.6 GB   3.4 GB   3.0 GB   3.9 GB   4.1 GB   3.9 GB   4.7 GB   4.8 GB   4.7 GB   5.6 GB   7.3 GB
Supported Languagesen ja
GGUF QuantizationYes
Quantization Typegguf
Model ArchitectureAutoModel
Licensellama2

Best Alternatives to Swallow 7B Instruct GGUF

Best Alternatives
Context / RAM
Downloads
Likes
Pixel8K / 4.4 GB170
...ral 7B Instruct V0.2 Llamafile0K / 14.5 GB885025
Mistral 7B Instruct V0.3 GGUF0K / 1.6 GB91691147
Qwen2 7B Instruct GGUF0K / 1.9 GB5641410
...hemeng Qwen Math 7b 24 1 100 10K / 15.2 GB250
Mistral 7B Instruct V0.2 GGUF0K / 3.1 GB70243511
Qwen2 7B Instruct V0.6 GGUF0K / 4.5 GB135220
Qwen2 7B Instruct V0.1 GGUF0K / 4.5 GB97140
Qwen2 7B Instruct V0.7 GGUF0K / 4.5 GB95300
Qwen2 7B Instruct V0.3 GGUF0K / 4.5 GB89111
Note: green Score (e.g. "73.2") means that the model is better than TheBloke/Swallow-7B-Instruct-GGUF.