What are the hardware requirements for ImagineGPT?

ImagineGPT requires approximately 0.5 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed ImagineGPT and how large is it?

ImagineGPT is developed by d-s-b, a model with 268.1m parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does ImagineGPT perform on standard benchmarks?

ImagineGPT has the following published scores: LMSYS ELO 1338. Compare against reference models on this page or on the LLM Explorer leaderboards.

ImagineGPT by d-s-b — VRAM 0.5GB, 32K context

Name: ImagineGPT
Author: d-s-b

ImagineGPT is an open-source language model by d-s-b. Features: 268.1m LLM, VRAM: 0.5GB, Context: 32K, LLM Explorer Score: 0.33, ELO: 1338.

Base model:d-s-b/imaginegpt Base model:finetune:d-s-b/imag... Conversational Endpoints compatible Gemma3 text Generated from trainer Region:us Safetensors Sft Tensorboard Trl

Model Card on HF 🤗: https://huggingface.co/d-s-b/ImagineGPT

ImagineGPT Benchmarks

LMSys ELO 1338 vs 1272 (so35) 5.2%

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

ImagineGPT Parameters and Internals

LLM Name	ImagineGPT
Repository 🤗	https://huggingface.co/d-s-b/ImagineGPT
Model Name	ImagineGPT
Base Model(s)	ImagineGPT d-s-b/ImagineGPT
Model Size	268.1m
Required VRAM	0.5 GB
Updated	2026-07-14
Maintainer	d-s-b
Model Type	gemma3_text
Model Files	0.5 GB 0.0 GB
Model Architecture	Gemma3ForCausalLM
Context Length	32768
Model Max Length	32768
Transformers Version	4.55.2
Tokenizer Class	GemmaTokenizer
Padding Token	<pad>
Vocabulary Size	262144
Torch Data Type	bfloat16

Rank the ImagineGPT 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

Best Alternatives to ImagineGPT

Best Alternatives	Context / RAM	Downloads
Episodic Nothink3b	32K / 0.5 GB	10
MyGemmaNPC	32K / 0.5 GB	15
Nacup	32K / 0.5 GB	9
Zer0 Gemma Rust Finetuned	32K / 0.5 GB	264