Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features: - Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios. - Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and…
Open-weight model · Text generation
qwen3-4b-base-dapo-v4
by Reliquary ReliquaryForge/qwen3-4b-base-dapo-v4
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.
Parameters4B
Context32,768
Weights8.0 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads910.6k
Runs On
What it takes to serve qwen3-4b-base-dapo-v4 (4B 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 | 8.0 GB | 9.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 4.0 GB | 4.8 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 2.0 GB | 2.4 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.
SAVRN's Notes on qwen3-4b-base-dapo-v4
Nothing on the hardware side of this one is hard. At 16-bit it wants 9.7 GB of memory, 8-bit takes 4.8 GB and 4-bit takes 2.4 GB, so the $1.85 an hour MI300X the Index lists as cheapest, with 192 GB, could hold a dozen copies. Four billion parameters and a 32,768 token context put it in the class we run for batch text work, one card serving many streams.
Apache 2.0 permits commercial use, modification and redistribution, and access is open. What a buyer should check is provenance. The name points at Qwen3-4B-Base with a DAPO pass, yet the page's only relation is the Qwen3 technical report, with no derived-from entry and no evaluations. The publisher text reads as the base model's description, not an account of what Reliquary changed. Ask for that account, and pin the revision you test, since the listing was last updated on 2026-09-18.
Model Card
By Reliquary, published under apache-2.0, revision ff8fc7e2c40e.
Qwen3-4B-Base
Qwen3 Highlights
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:
Configuration
- Architecture
- Qwen3ForCausalLM
- Context length (tokens)
- 32,768
- Layers
- 36
- Hidden size
- 2,560
- Feed-forward size
- 9,728
- Attention heads
- 32
- Key/value heads
- 8
- Head dimension
- 128
- Vocabulary size
- 151,936
- Model type
- qwen3
Identity and Version
- Repository
- ReliquaryForge/qwen3-4b-base-dapo-v4
- Publisher
- Reliquary
- Task
- Text generation
- Modality
- Text
- Library
- transformers
- Parameters
- 4B parameters
- Languages
- Not stated by the source
- Revision
- ff8fc7e2c40ee0b83f820dd23f5249afeb0e44f9
- First published
- 2026-08-18
- Last updated
- 2026-09-18
Files and Weights
14 files, 8.1 GB in total. The weights are 1 file totalling 8.0 GB in safetensors.
Weights1 file · 8.0 GB
Configuration5 files · 5.6 KB
Tokenizer4 files · 15.9 MB
Documentation2 files · 14.3 KB
Other1 file · 4.1 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 8.0 GB | baf59a867a81 |
| config.json | Configuration | 1.6 KB | — |
| generation_config.json | Configuration | 139 B | — |
| reliquary_protocol_profile.json | Configuration | 437 B | — |
| reliquary_publication.json | Configuration | 1.7 KB | — |
| reliquary_v1_bootstrap.json | Configuration | 1.7 KB | — |
| LICENSE | Documentation | 11.3 KB | — |
| README.md | Documentation | 2.9 KB | — |
| chat_template.jinja | Other | 4.1 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| merges.txt | Tokenizer | 1.7 MB | — |
| tokenizer.json | Tokenizer | 11.4 MB | be75606093db |
| tokenizer_config.json | Tokenizer | 696 B | — |
| vocab.json | Tokenizer | 2.8 MB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 8.0 GB
Download from Reliquary
Released by Reliquary through its official repository on Hugging Face. Read the license.
Built From
- Described by arXiv:2505.09388
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 8.0 GB |
| 16-bit | 8.0 GB |
| 8-bit | 4.0 GB |
| 4-bit | 2.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Compare qwen3-4b-base-dapo-v4
Questions About qwen3-4b-base-dapo-v4
How much GPU memory does qwen3-4b-base-dapo-v4 need?
About 9.7 GB at 16-bit and 2.4 GB at 4-bit: the weights (4B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run qwen3-4b-base-dapo-v4 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 qwen3-4b-base-dapo-v4 commercially?
Yes. qwen3-4b-base-dapo-v4 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 qwen3-4b-base-dapo-v4's context length?
32,768 tokens, from the maximum position embeddings in its published configuration.
Similar Models
We introduce the updated version of the Qwen3-4B non-thinking mode, named Qwen3-4B-Instruct-2507, featuring the following key enhancements: - Significant improvements in general capabilities, including instruction following, logical reasoning, text comprehension, mathematics, science, coding and tool usage. - Substantial gains in long-tail knowledge coverage across multiple languages. - Markedly better alignment with user preferences in subjective and open-ended tasks, enabling more helpful responses and higher-quality text generation. - Enhanced capabilities in 256K long-context understanding. Qwen3-4B-Instruct-2507 has the following features: NOTE: This model supports only non-thinking…
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: Qwen3-4B-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…
We introduce the updated version of the Qwen3-4B non-thinking mode, named Qwen3-4B-Instruct-2507, featuring the following key enhancements: - Significant improvements in general capabilities, including instruction following, logical reasoning, text comprehension, mathematics, science, coding and tool usage. - Substantial gains in long-tail knowledge coverage across multiple languages. - Markedly better alignment with user preferences in subjective and open-ended tasks, enabling more helpful responses and higher-quality text generation. - Enhanced capabilities in 256K long-context understanding. Qwen3-4B-Instruct-2507 has the following features: NOTE: This model supports only non-thinking…
This model is a fine-tuned version of cosmos1030/gmp-kd3e-1-s80pct-lr1e-420260916220740 on the trl-lib/ultrafeedbackbinarized dataset. It has been trained using TRL. This model was trained with DPO, a method introduced in Direct Preference Optimization: Your Language Model is Secretly a Reward Model.
The full ImmuneCoT method fuses the two safety branches with a Base-adjusted product-of-experts Qimm(v) ∝ q̃rec(v)·q̃res(v)/qB(v). This checkpoint uses the naive fusion Qno-base(v) ∝ q̃rec(v)·q̃res(v) — the same branch weights (λrec=0.5, λres=0.7) but no division by the base distribution — isolating whether the gains come from combining Recognition+Response at all, or specifically from the Base-adjusted PoE term. Intended use: research reproducibility for the ImmuneCoT paper's RQ3 ablation.