What are the hardware requirements for TinyBalls V1 Base?

TinyBalls V1 Base requires approximately 0.5 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed TinyBalls V1 Base and how large is it?

TinyBalls V1 Base is developed by igidn, a model with 113.3m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate TinyBalls V1 Base?

TinyBalls V1 Base is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

TinyBalls V1 Base by igidn — VRAM 0.5GB, 2K context

Name: TinyBalls V1 Base
Author: igidn

TinyBalls V1 Base is an open-source language model by igidn. Features: 113.3m LLM, VRAM: 0.5GB, Context: 2K, License: mit, LLM Explorer Score: 0.31.

Dataset:codeparrot/codeparrot-... Dataset:common-pile/biodiversi... Dataset:common-pile/caselaw ac... Dataset:common-pile/doab Dataset:common-pile/library of... Dataset:common-pile/libretexts... Dataset:common-pile/oercommons... Dataset:common-pile/pes2o Dataset:common-pile/pre 1929 b... Dataset:common-pile/pressbooks... Dataset:common-pile/project gu... Dataset:common-pile/python enh... Dataset:common-pile/regulation... Dataset:common-pile/stackexcha... Dataset:common-pile/stackexcha... Dataset:common-pile/ubuntu irc Dataset:common-pile/uspto Dataset:common-pile/youtube Dataset:eleutherai/proof-pile-... Dataset:glaiveai/glaive-functi... Dataset:huggingfacefw/fineweb-... Dataset:huggingfacefw/finewiki Dataset:huggingfacetb/cosmoped... Dataset:huggingfacetb/finemath Dataset:nebius/swe-agent-traje... Dataset:nvidia/opencodereasoni... Dataset:open-web-math/open-web... Dataset:opencoder-llm/opc-anne... Dataset:swe-gym/openhands-samp... Dataset:tiger-lab/webinstructs... En Llama Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/igidn/TinyBalls-v1-Base

TinyBalls V1 Base Parameters and Internals

LLM Name	TinyBalls V1 Base
Repository 🤗	https://huggingface.co/igidn/TinyBalls-v1-Base
Model Size	113.3m
Required VRAM	0.5 GB
Updated	2026-10-07
Maintainer	igidn
Model Type	llama
Model Files	0.5 GB
Supported Languages	en
Model Architecture	LlamaForCausalLM
License	mit
Context Length	2048
Model Max Length	2048
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<\|endoftext\|>
Vocabulary Size	49154
Torch Data Type	float32

Rank the TinyBalls V1 Base capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

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