LLM EXPLORER 59,420 MODELS INDEXED

Akins 3B by acrastt

By acrastt · 14 downloads

Akins 3B is an open-source language model by acrastt. Features: 3b LLM, VRAM: 5.6GB, Context: 4K, License: cc-by-sa-4.0, LLM Explorer Score: 0.09.

  Arxiv:2305.14314   Custom code Dataset:norquinal/claude multi...   En   Pytorch   Region:us   Safetensors   Stablelm epoch
Model Card on HF πŸ€—: https://huggingface.co/acrastt/Akins-3B 

Akins 3B Parameters and Internals

Model Type 
text generation
Additional Notes 
GGUF quantizations available [here](https://huggingface.co/TheBloke/Akins-3B-GGUF). GPTQ quantizations available [here](https://huggingface.co/TheBloke/Akins-3B-GPTQ).
Supported Languages 
en (English)
Training Details 
Data Sources:
Norquinal/claude_multiround_chat_1k
Methodology:
instruction-tuned with QLoRA
Training Time:
2 epochs
Input Output 
Input Format:
USER: {prompt} ASSISTANT:
Accepted Modalities:
text
LLM NameAkins 3B
Repository πŸ€—https://huggingface.co/acrastt/Akins-3B 
Model Size3b
Required VRAM5.6 GB
Updated2026-08-01
Maintaineracrastt
Model Typestablelm_epoch
Model Files  5.6 GB   5.6 GB
Supported Languagesen
Model ArchitectureStableLMEpochForCausalLM
Licensecc-by-sa-4.0
Context Length4096
Model Max Length4096
Transformers Version4.35.0.dev0
Tokenizer ClassGPTNeoXTokenizerFast
Vocabulary Size50304
Torch Data Typefloat16

Quantized Models of the Akins 3B

Model
Likes
Downloads
VRAM
Akins 3B GGUF82141 GB
Akins 3B GPTQ261 GB

Best Alternatives to Akins 3B

Best Alternatives
Context / RAM
Downloads
Likes
Stable Code 3B Mlx16K / 5.6 GB171
Aura 3B4K / 5.6 GB32
Slim Extract4K / 5.6 GB1013
Slim Boolean4K / 5.6 GB94
Slim Sa Ner4K / 5.6 GB206
Slim Tags 3B4K / 5.6 GB114
Slim Summary4K / 5.6 GB108
Slim Xsum4K / 5.6 GB146
Tofu 3B4K / 5.6 GB52
Memphis CoT 3B4K / 5.6 GB2431
Note: green Score (e.g. "73.2") means that the model is better than acrastt/Akins-3B.