LLM EXPLORER 59,516 MODELS INDEXED

GPT Sw3 20B Instruct by AI-Sweden-Models

By AI-Sweden-Models · 1129 downloads
HF Repo ↗ Copy URL

GPT Sw3 20B Instruct is an open-source language model by AI-Sweden-Models. Features: 20b LLM, VRAM: 83.2GB, License: other, Instruction-Based, LLM Explorer Score: 0.11, Arc: 43.2, HellaSwag: 71.1, MMLU: 31.3, GSM8K: 8.8.

Base model:ai-sweden-models/gp... Base model:finetune:ai-sweden-...   Conversational   Da Dataset:databricks/databricks-...   Dataset:laion/oig   Dataset:openassistant/oasst1   Deploy:azure   En   Endpoints compatible   Gpt2   Instruct   Is   No   Pytorch   Region:us   Safetensors   Sharded   Sv   Tensorflow
Model Card on HF 🤗: https://huggingface.co/AI-Sweden-Models/gpt-sw3-20b-instruct 

GPT Sw3 20B Instruct Parameters and Internals

LLM NameGPT Sw3 20B Instruct
Repository 🤗https://huggingface.co/AI-Sweden-Models/gpt-sw3-20b-instruct 
Base Model(s)  GPT Sw3 20B   AI-Sweden-Models/gpt-sw3-20b
Model Size20b
Required VRAM83.2 GB
Updated2026-05-28
MaintainerAI-Sweden-Models
Model Typegpt2
Instruction-BasedYes
Model Files  9.5 GB: 1-of-9   9.7 GB: 2-of-9   9.7 GB: 3-of-9   9.7 GB: 4-of-9   9.7 GB: 5-of-9   9.7 GB: 6-of-9   9.7 GB: 7-of-9   9.7 GB: 8-of-9   5.8 GB: 9-of-9   9.5 GB: 1-of-9   9.7 GB: 2-of-9   9.7 GB: 3-of-9   9.7 GB: 4-of-9   9.7 GB: 5-of-9   9.7 GB: 6-of-9   9.7 GB: 7-of-9   9.7 GB: 8-of-9   5.8 GB: 9-of-9
Supported Languagesda sv en no is
Model ArchitectureGPT2LMHeadModel
Licenseother
Model Max Length2048
Transformers Version4.22.1
Tokenizer ClassGPTSw3Tokenizer
Padding Token<pad>
Vocabulary Size64000
Torch Data Typefloat32
Activation Functiongelu

Best Alternatives to GPT Sw3 20B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
Palmyra 20B Chat0K / 40.6 GB17311
Note: green Score (e.g. "73.2") means that the model is better than AI-Sweden-Models/gpt-sw3-20b-instruct.