What are the hardware requirements for SFT DART?

SFT DART requires approximately 1.2 GB of VRAM and supports a context window of 40K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed SFT DART and how large is it?

SFT DART is developed by Cusul, a model with 0.6b 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 SFT DART?

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

SFT DART by Cusul — VRAM 1.2GB, 40K context

Name: SFT DART
Author: Cusul

SFT DART is an open-source language model by Cusul. Features: 0.6b LLM, VRAM: 1.2GB, Context: 40K, LLM Explorer Score: 0.18.

Base model:finetune:qwen/qwen3... Base model:qwen/qwen3-0.6b Conversational Endpoints compatible Generated from trainer Qwen3 Region:us Safetensors Sft Trl

Model Card on HF 🤗: https://huggingface.co/Cusul/SFT_DART

SFT DART Parameters and Internals

LLM Name	SFT DART
Repository 🤗	https://huggingface.co/Cusul/SFT_DART
Model Name	SFT_DART
Base Model(s)	Qwen/Qwen3-0.6B Qwen/Qwen3-0.6B
Model Size	0.6b
Required VRAM	1.2 GB
Updated	2026-10-08
Maintainer	Cusul
Model Type	qwen3
Model Files	1.2 GB 0.0 GB
Model Architecture	Qwen3ForCausalLM
Context Length	40960
Model Max Length	40960
Transformers Version	4.52.4
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Torch Data Type	bfloat16
Errors	replace

Rank the SFT DART 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 SFT DART

Best Alternatives	Context / RAM	Downloads	Likes
S1 Mini	40K / 1.5 GB	1136	180
Qwen3 Reranker 0.6B	40K / 1.2 GB	2508313	377
E2etext T1edu1 T1	40K / 1.2 GB	995	1
Qwen3 Finetuned	40K / 2.4 GB	178	1
Cultist Simulator Qwen3 0.6B	40K / 1.2 GB	297	0
Qwen3 0.6B JSON SFT	40K / 1.2 GB	231	0
Qwen3 0.6B JSON SFT	40K / 1.2 GB	231	0
Adtc Agri Sft Qwen3 0.6B	40K / 1.2 GB	619	0
Qwen3 0.6B Heretic	40K / 1.2 GB	238	0
Qwen3 0.6B FP8	40K / 1.1 GB	1581403	62

Note: green Score (e.g. "73.2") means that the model is better than Cusul/SFT_DART.