dQwen3-1.7B-Base by IFML: Open-Weight Model

Runs On

What it takes to serve dQwen3-1.7B-Base (1.7B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

Precision	Weights	Memory needed	Cheapest setup	Per hour	Also fits
16-bit	3.4 GB	4.1 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00
8-bit	1.7 GB	2.1 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00
4-bit	0.9 GB	1.0 GB	1x MI300X (192 GB) Vultr	$1.85	1x H100 $1.99 · 1x MI325X $2.00

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.

Model Card

By IFML, published under apache-2.0, revision 9a9cb6cb6995.

A masked diffusion language model adapted from Qwen3-1.7B. The backbone is full attention, and every layer is made bidirectional. It is the control model in the paper's matched comparison against the hybrid dQwen3.5-2B. This is a base model, with no instruction tuning. Paper: dQwen3.5: Hybrid-Attention Diffusion Language Models. Code: https://github.com/AntonXue/dQwen Needs a CUDA GPU and transformers>=5.13 (tested with torch 2.7.1+cu128, flash-linear-attention 0.5.1). generate decodes the whole canvas at once, committing positions above a confidence threshold (tau=0.9); pass blocklength=32 for left-to-right block decoding, or tau=None, stepsperblock=k for a fixed budget. The 50B-token…

Read IFML's full model card

A masked diffusion language model adapted from Qwen3-1.7B. The backbone is full attention, and every layer is made bidirectional. It is the control model in the paper's matched comparison against the hybrid dQwen3.5-2B. This is a base model, with no instruction tuning.

Paper: dQwen3.5: Hybrid-Attention Diffusion Language Models. Code: https://github.com/AntonXue/dQwen

The dQwen3.5 family

model	total params	backbone
dQwen3.5-0.8B-Base	0.75B	hybrid
dQwen3.5-2B-Base	1.88B	hybrid
dQwen3.5-4B-Base	4.21B	hybrid
dQwen3.5-9B-Base	8.95B	hybrid
dQwen3-1.7B-Base (this repo)	1.72B	full attention (control)

Quickstart

Needs a CUDA GPU and transformers>=5.13 (tested with torch 2.7.1+cu128, flash-linear-attention 0.5.1).

import torch
from transformers import AutoModel

model = AutoModel.from_pretrained("UT-IFML/dQwen3-1.7B-Base", trust_remote_code=True, dtype=torch.bfloat16).cuda().eval()
print(model.generate("def fibonacci(n):", gen_length=512, stop_strings=["\ndef "]).text)

generate decodes the whole canvas at once, committing positions above a confidence threshold (tau=0.9); pass block_length=32 for left-to-right block decoding, or tau=None, steps_per_block=k for a fixed budget. The 50B-token checkpoint from the paper is revision="step25000-swa".

Citation

@article{xue2026dqwen,
  title   = {dQwen3.5: Hybrid-Attention Diffusion Language Models},
  author  = {Xue, Anton and Rout, Litu and Akella, Aditya and Klivans, Adam and Sanghavi, Sujay and Shakkottai, Sanjay},
  journal = {arXiv preprint arXiv:2609.20751},
  year    = {2026}
}

Configuration

Architecture: DiffuQwen3
Context length (tokens): 40,960
Layers: 28
Hidden size: 2,048
Feed-forward size: 6,144
Attention heads: 16
Key/value heads: 8
Head dimension: 128
Vocabulary size: 151,936
Model type: diffuqwen3

Identity and Version

Repository: UT-IFML/dQwen3-1.7B-Base
Publisher: IFML
Task: Text generation
Modality: Text
Library: transformers
Parameters: 1.7B parameters
Languages: Not stated by the source
Revision: 9a9cb6cb69955407644869208236e4581fc56f11
First published: 2026-08-09
Last updated: 2026-09-18

Files and Weights

11 files, 3.5 GB in total. The weights are 1 file totalling 3.4 GB in safetensors.

Weights1 file · 3.4 GB

Configuration4 files · 16.2 KB

Tokenizer2 files · 11.4 MB

Documentation1 file · 2.2 KB

Other2 files · 1.1 MB

Repository1 file · 1.6 KB

Every file

File	Type	Size	SHA-256
model.safetensors	Weights	3.4 GB	0fdfbf458c53
argparse.json	Configuration	870 B	—
config.json	Configuration	1.5 KB	—
generation_config.json	Configuration	205 B	—
modeling_dqwen3.py	Configuration	13.6 KB	—
README.md	Documentation	2.2 KB	—
chat_template.jinja	Other	4.2 KB	—
training_log.jsonl	Other	1.1 MB	—
.gitattributes	Repository	1.6 KB	—
tokenizer.json	Tokenizer	11.4 MB	be75606093db
tokenizer_config.json	Tokenizer	745 B	—

License and Download

License: apache-2.0
Access: Open weights, no gate
Download size: 3.4 GB

Download from IFML

Released by IFML through its official repository on Hugging Face. Read the license.

Built From

Derived from Qwen/Qwen3-1.7B
Described by arXiv:2609.20751

Memory Requirements

Precision	Weights in memory
As published	3.4 GB
16-bit	3.4 GB
8-bit	1.7 GB
4-bit	0.9 GB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About dQwen3-1.7B-Base

How much GPU memory does dQwen3-1.7B-Base need?

About 4.1 GB at 16-bit and 1 GB at 4-bit: the weights (1.7B parameters) plus a working margin. A long context needs more.

What is the cheapest GPU to run dQwen3-1.7B-Base on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use dQwen3-1.7B-Base commercially?

Yes. dQwen3-1.7B-Base is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

What is dQwen3-1.7B-Base's context length?

40,960 tokens, from the maximum position embeddings in its published configuration.

Similar Models

Model · Text generation

Qwen3-1.7B-Base

Qwen

Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Building upon extensive advancements in training data, model architecture, and optimization techniques, Qwen3 delivers the following key improvements over the previously released Qwen2.5: Qwen3-1.7B-Base has the following features: For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog, GitHub, and Documentation. The code of Qwen3 has been in the latest Hugging Face transformers and we advise you to use the latest version of transformers. With transformers<4.51.0, you will…

Open weights apache-2.0 1.7B parameters 32,768 tokens transformers

View model

Model · Text generation

DualMind-TKD-Agentic-1.7B

Convergent Intelligence

DualMind TKD Agentic 1.7B is a two-stage derivative of Qwen/Qwen3-1.7B. It combines topology-guided mathematical knowledge distillation with assistant-masked agentic and function-calling specialization. teacher distillation topology, gap-energy diagnostics, and phase-weighted Explore/Examine/Response supervision Stage 1 was designed to transfer mathematical reasoning behavior while placing additional learning pressure on derivation, verification, and high-discrepancy reasoning transitions. - Tool schemas, user messages, and tool-result messages were visible as context but excluded from direct loss - Mathematical replay was mixed into Stage 2 to reduce catastrophic forgetting The files in…

Open weights other 1.7B parameters 40,960 tokens transformers

View model

Model · Text generation

qwen3-1.7b-grpo

Yu

This model is a fine-tuned version of None. It has been trained using TRL. This model was trained with GRPO, a method introduced in DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models.

Open weights 1.7B parameters 40,960 tokens transformers

View model

Model · Text generation

Shepherd-Alpha

Convergent Intelligence

The first defense AI reasoning model on Hugging Face. Shepherd-Alpha is a tactical reasoning model fine-tuned on dual-perspective military scenario analysis using BiCell Depth Dispersal — a novel training methodology that partitions transformer layers by abstraction depth and trains them asymmetrically to separate representation encoding from task-specific reasoning. Developed by Convergent Intelligence LLC: Research Division Given a tactical scenario, Shepherd-Alpha produces structured dual-perspective analysis: - Attack reasoning — how an adversary would exploit the situation - Defense reasoning — how to counter, mitigate, and survive The model is trained to think like both attacker and…

Open weights apache-2.0 1.7B parameters 40,960 tokens transformers

View model

Model · Text generation

Bonsai-27B-mlx-1bit

Prism ML

Full 27B-class reasoning in binary transformer weights — the first 27B-class model to run on a phone - ~3.9 GB deployed footprint (down from ~54 GB FP16) — fits within the per-app memory budget of a high-end phone such as the iPhone 17 Pro Max - Retains thinking, reasoning, and agentic behavior deep in the sub-4-bit regime, where conventional low-bit representations collapse — 76.11 average across 15 thinking-mode benchmarks (89.5% of FP16), including math at 91.66 and coding at 81.88 - End-to-end binary language weights across embeddings, attention projections, MLP projections, and LM head, at a true 1.125 bits per weight — no high-precision escape hatches behind a low-bit label; the…

Open weights apache-2.0 1.7B parameters 262,144 tokens mlx

View model

Model · Text generation

JiRackUltra_1b

Center Business Solutions inc

A fast and efficient ~1.5B model optimized for CPU inference. The model was refactored with BitNet features and an updated tokenizer that includes new Routing, Tool call, and Robotics tags. Built on a redesigned DeepSeek R1 architecture with native ternary (BitNet-style) support and ready-to-run GGUF quantizations. - JiRack is a cloud-ready model that helps save money on cloud infrastructure. It can be used as an expert model in RAG deployments, with the ONNX JiRack Java server as an alternative. - Benefits high quality CPU inference TQ2 on Llama.cpp and Ollama via QAT - Robotcs, Routing, Coding, Multimedia, Advanced tool calling via CMSManhattan/JiRackPrecisionTokenizer - We are working to…

Open weights mit 1.8B parameters 131,072 tokens

View model