LLM EXPLORER 57,252 MODELS INDEXED

Apertus 70B Instruct 2509 Unsloth Bnb 4bit by unsloth

By unsloth · 48 downloads
HF Repo ↗ Copy URL

Apertus 70B Instruct 2509 Unsloth Bnb 4bit is an open-source language model by unsloth. Features: 70b LLM, VRAM: 115GB, Context: 64K, License: apache-2.0, Quantized, Instruction-Based, LLM Explorer Score: 0.2.

  Arxiv:2509.14233   4-bit   4bit   Apertus Base model:quantized:swiss-ai/... Base model:swiss-ai/apertus-70...   Bitsandbytes   Compliant   Conversational   Endpoints compatible   Instruct   Multilingual   Quantized   Region:us   Safetensors   Sharded   Swiss-ai   Tensorflow
Model Card on HF 🤗: https://huggingface.co/unsloth/Apertus-70B-Instruct-2509-unsloth-bnb-4bit 

Apertus 70B Instruct 2509 Unsloth Bnb 4bit Parameters and Internals

LLM NameApertus 70B Instruct 2509 Unsloth Bnb 4bit
Repository 🤗https://huggingface.co/unsloth/Apertus-70B-Instruct-2509-unsloth-bnb-4bit 
Base Model(s)  swiss-ai/Apertus-70B-2509   swiss-ai/Apertus-70B-2509
Model Size70b
Required VRAM115 GB
Updated2026-07-28
Maintainerunsloth
Model Typeapertus
Instruction-BasedYes
Model Files  4.9 GB: 1-of-25   4.4 GB: 2-of-25   5.0 GB: 3-of-25   5.0 GB: 4-of-25   4.9 GB: 5-of-25   4.8 GB: 6-of-25   4.6 GB: 7-of-25   4.5 GB: 8-of-25   4.6 GB: 9-of-25   4.6 GB: 10-of-25   4.6 GB: 11-of-25   4.7 GB: 12-of-25   4.6 GB: 13-of-25   4.5 GB: 14-of-25   4.5 GB: 15-of-25   4.6 GB: 16-of-25   4.5 GB: 17-of-25   4.6 GB: 18-of-25   4.5 GB: 19-of-25   4.5 GB: 20-of-25   4.6 GB: 21-of-25   4.9 GB: 22-of-25   5.0 GB: 23-of-25   5.0 GB: 24-of-25   2.6 GB: 25-of-25
Quantization Type4bit
Model ArchitectureApertusForCausalLM
Licenseapache-2.0
Context Length65536
Model Max Length65536
Transformers Version4.56.2
Tokenizer ClassPreTrainedTokenizerFast
Padding Token<pad>
Vocabulary Size131072
Torch Data Typebfloat16