What are the hardware requirements for NanoLlama GQA L10 A32 KV8 V13 KI?

NanoLlama GQA L10 A32 KV8 V13 KI requires approximately 0.4 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 NanoLlama GQA L10 A32 KV8 V13 KI and how large is it?

NanoLlama GQA L10 A32 KV8 V13 KI is developed by BEE-spoke-data, a model with 217.9m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does NanoLlama GQA L10 A32 KV8 V13 KI perform on standard benchmarks?

NanoLlama GQA L10 A32 KV8 V13 KI has the following published scores: MMLU 25.37. Compare against reference models on this page or on the LLM Explorer leaderboards.

NanoLlama GQA L10 A32 KV8 V13 KI by BEE-spoke-data — VRAM 0.4GB, 2K context

Name: NanoLlama GQA L10 A32 KV8 V13 KI
Author: BEE-spoke-data

NanoLlama GQA L10 A32 KV8 V13 KI is an open-source language model by BEE-spoke-data. Features: 217.9m LLM, VRAM: 0.4GB, Context: 2K, License: apache-2.0, LLM Explorer Score: 0.15, Arc: 23.8, HellaSwag: 29.4, MMLU: 25.4, GSM8K: 0.9.

Endpoints compatible Generated from trainer Llama Llama2 Model-index Region:us Safetensors Smol llama

Model Card on HF 🤗: https://huggingface.co/BEE-spoke-data/NanoLlama-GQA-L10-A32_KV8-v13-KI

NanoLlama GQA L10 A32 KV8 V13 KI Parameters and Internals

Model Type

text-generation

Additional Notes

The model is a fine-tuned version of the base model, with training still a work in progress. It achieves a loss of 2.5937 and an accuracy of 0.4948 on the evaluation set.

Training Details

Data Sources:

KI dataset

LLM Name	NanoLlama GQA L10 A32 KV8 V13 KI
Repository 🤗	https://huggingface.co/BEE-spoke-data/NanoLlama-GQA-L10-A32_KV8-v13-KI
Base Model(s)	BEE-spoke-data/NanoLlama-GQA-L10-A32_KV8-v12-minipile BEE-spoke-data/NanoLlama-GQA-L10-A32_KV8-v12-minipile
Model Size	217.9m
Required VRAM	0.4 GB
Updated	2026-08-13
Maintainer	BEE-spoke-data
Model Type	llama
Model Files	0.4 GB 0.0 GB
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	2048
Model Max Length	2048
Transformers Version	4.36.0.dev0
Tokenizer Class	LlamaTokenizer
Vocabulary Size	32128
Torch Data Type	bfloat16

Rank the NanoLlama GQA L10 A32 KV8 V13 KI 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