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
DeepSeek-R1-0528-Qwen3-8B
by DeepSeek deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
DeepSeek-R1-0528-Qwen3-8B is an open-weight model for text generation from DeepSeek, released under MIT License. It has 8.2B parameters and a 131,072-token context. At 16-bit it needs about 19.7 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index. It draws 895.6k downloads a month.
The DeepSeek R1 model has undergone a minor version upgrade, with the current version being DeepSeek-R1-0528.
Parameters8.2B
Context131,072
Weights16.4 GB
Licensemit
AccessOpen weights
Monthly Downloads895.6k
Runs On
What it takes to serve DeepSeek-R1-0528-Qwen3-8B (8.2B 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 | 16.4 GB | 19.7 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 8.2 GB | 9.8 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 4.1 GB | 4.9 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 Oct 13, 2026.
DeepSeek-R1-0528-Qwen3-8B on every accelerator the SAVRN Index prices, at every precision
SAVRN's Notes on DeepSeek-R1-0528-Qwen3-8B
At 16-bit precision this model asks for 19.7 GB of memory, which settles the hardware question early. DeepSeek built it on the Qwen3ForCausalLM architecture at 8.2 billion parameters for text generation, with a 131,072-token context window and weights that occupy 16.4 GB as safetensors. Drop to 8-bit and the memory need falls to 9.8 GB; at 4-bit it is 4.9 GB. The cheapest setup on our Index is a single MI300X with 192 GB at $1.85 per hour on demand, so one card holds it many times over.
The MIT license permits commercial use, modification and redistribution as long as the copyright and permission notices stay with the files. Before committing, confirm your workload needs the full 131,072-token context, read the describing paper arXiv:2501.12948, and note that our Index lists no per-token host prices for this model yet, so the hourly card rate is the only cost reference.
Model Card
By DeepSeek, published under mit, revision 6e8885a6ff5c.
DeepSeek-R1-0528
1. Introduction
The DeepSeek R1 model has undergone a minor version upgrade, with the current version being DeepSeek-R1-0528. In the latest update, DeepSeek R1 has significantly improved its depth of reasoning and inference capabilities by leveraging increased computational resources and introducing algorithmic optimization mechanisms during post-training. The model has demonstrated outstanding performance across various benchmark evaluations, including mathematics, programming, and general logic. Its overall performance is now approaching that of leading models, such as O3 and Gemini 2.5 Pro.
Compared to the previous version, the upgraded model shows significant improvements in handling complex reasoning tasks. For instance, in the AIME 2025 test, the model’s accuracy has increased from 70% in the previous version to 87.5% in the current version. This advancement stems from enhanced thinking depth during the reasoning process: in the AIME test set, the previous model used an average of 12K tokens per question, whereas the new version averages 23K tokens per question.
Configuration
- Architecture
- Qwen3ForCausalLM
- Context length (tokens)
- 131,072
- Layers
- 36
- Hidden size
- 4,096
- Feed-forward size
- 12,288
- Attention heads
- 32
- Key/value heads
- 8
- Head dimension
- 128
- Vocabulary size
- 151,936
- RoPE base
- 1,000,000
- Stored precision
- bfloat16
- Model type
- qwen3
Identity and Version
- Repository
- deepseek-ai/DeepSeek-R1-0528-Qwen3-8B
- Publisher
- DeepSeek
- Task
- Text generation
- Modality
- Text
- Library
- transformers
- Parameters
- 8.2B parameters
- Languages
- Not stated by the source
- Revision
- 6e8885a6ff5c1dc5201574c8fd700323f23c25fa
- First published
- 2025-05-29
- Last updated
- 2025-05-29
Files and Weights
10 files, 16.4 GB in total. The weights are 2 files totalling 16.4 GB in safetensors.
Weights2 files · 16.4 GB
Configuration2 files · 34.1 KB
Tokenizer2 files · 7.0 MB
Documentation2 files · 16.0 KB
Other1 file · 322.5 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-000002.safetensors | Weights | 8.6 GB | 7bb656406ecf |
| model-00002-of-000002.safetensors | Weights | 7.8 GB | 8c3631d826c0 |
| config.json | Configuration | 859 B | — |
| model.safetensors.index.json | Configuration | 33.3 KB | — |
| LICENSE | Documentation | 1.1 KB | — |
| README.md | Documentation | 14.9 KB | — |
| figures/benchmark.png | Other | 322.5 KB | 60aba4b9eb56 |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 7.0 MB | — |
| tokenizer_config.json | Tokenizer | 4.0 KB | — |
License and Download
- License
- mit
- Access
- Open weights, no gate
- Download size
- 16.4 GB
Download from DeepSeek
Released by DeepSeek through its official repository on Hugging Face. Read the license.
Built From
- Described by arXiv:2501.12948
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 16.4 GB |
| 16-bit | 16.4 GB |
| 8-bit | 8.2 GB |
| 4-bit | 4.1 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
Compare DeepSeek-R1-0528-Qwen3-8B
Questions About DeepSeek-R1-0528-Qwen3-8B
How much GPU memory does DeepSeek-R1-0528-Qwen3-8B need?
About 19.7 GB at 16-bit and 4.9 GB at 4-bit: the weights (8.2B parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run DeepSeek-R1-0528-Qwen3-8B 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 DeepSeek-R1-0528-Qwen3-8B commercially?
Yes. DeepSeek-R1-0528-Qwen3-8B is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.
What is DeepSeek-R1-0528-Qwen3-8B's context length?
131,072 tokens, from the maximum position embeddings in its published configuration.
Similar Models
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…
Purpose: Long-memory symbolic reasoning + high-fidelity language generation SymbioticLM-8B is a state-of-the-art hybrid transformer model with built-in symbolic cognition. It combines an 8B Qwen-based transformer with modular symbolic processors and a persistent memory buffer. The model supports both general conversation and deep symbolic tasks such as theorem generation, logical chaining, and structured reasoning with retained memory across turns. - General symbolic reasoning and logical conversation - Code + math proof modeling - Not instruction-tuned (e.g., chat-style inputs may require prompt engineering) - Larger memory buffer may increase CPU load slightly - Symbolic inference is…
English | 简体中文 Qwen3-8B-CC-SFT-v2 is a failure-targeted continued-SFT checkpoint designed to improve coding-agent state preservation and continuation across Claude Code native context compaction. It is initialized from Qwen3-8B-CC-SFT-v1 and trained on compact-aware, context-correct supervision distilled from GLM-5.3 under real Claude Code native compaction. Best observed canonical303 Pass@1: 52.48% (159/303) — 4090 run of this same checkpoint at T=0.5 (temperature selected on an independent dev set; see the table below). Best observed configurations of this same checkpoint: - T=0.5 was selected on an independent 60-task development set (task-id and repo disjoint from canonical303), then…
Ido Levy 1,2 · 1,2 The trained questioner from Asking for What Was Never Requested: Horizontal and Vertical Proactivity in Agents, with its LoRA adapter merged into Qwen3-8B. It is a standard full-weight model: it loads without PEFT and serves with vLLM, SGLang or TGI like any Qwen3-8B. - The adapter, with the results, the training details, the limitations and a complete two-turn - Quantized for llama.cpp, Ollama and LM Studio: This is training seed 1, the adapter at the root of the adapter repository. The merge ran in float32 and the weights are stored in bfloat16. On the adapter card's two-turn example, greedy decoding with this model returns the adapter's output character for character.…
