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
Socio-Foundation-8B
by Wenxuan Xie(SII) SII-LancelotXie/Socio-Foundation-8B
Socio-Foundation-8B is an open-weight model for text generation from Wenxuan Xie(SII), released under Apache License 2.0. It has 8.2B parameters and a 40,960-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 144 downloads a month.
Socio-Foundation-8B 是一个采用 Qwen3 架构的因果语言模型。本仓库提供完整的 BF16 模型权重、配置文件、分词器和聊天模板，可通过 Hugging Face Transformers 加载。 作者：Liang Wang, Wenxuan Xie, Xinyi Mou, Yixin Luo, Zhongyu Wei。 训练方法、FONTS 能力体系和 IndiEval 评测结果详见论文。 Apache-2.0。
Parameters8.2B
Context40,960
Weights16.4 GB
Licenseapache-2.0
AccessOpen weights
Monthly Downloads144
Runs On
What it takes to serve Socio-Foundation-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 9, 2026.
Socio-Foundation-8B on every accelerator the SAVRN Index prices, at every precision
Model Card
By Wenxuan Xie(SII), published under apache-2.0, revision 537c97d78ee8.
Socio-Foundation-8B 是一个采用 Qwen3 架构的因果语言模型。本仓库提供完整的 BF16 模型权重、配置文件、分词器和聊天模板，可通过 Hugging Face Transformers 加载。 作者：Liang Wang, Wenxuan Xie, Xinyi Mou, Yixin Luo, Zhongyu Wei。 训练方法、FONTS 能力体系和 IndiEval 评测结果详见论文。 Apache-2.0。
Read Wenxuan Xie(SII)'s full model card
Socio-Foundation-8B 是一个采用 Qwen3 架构的因果语言模型。本仓库提供完整的 BF16 模型权重、配置文件、分词器和聊天模板，可通过 Hugging Face Transformers 加载。
论文
作者：Liang Wang, Wenxuan Xie, Xinyi Mou, Yixin Luo, Zhongyu Wei。
训练方法、FONTS 能力体系和 IndiEval 评测结果详见论文。
模型信息
| 项目 | 说明 |
|---|---|
| 模型架构 | Qwen3ForCausalLM |
| 参数量 | 约 8.19B |
| Transformer 层数 | 36 |
| 权重精度 | BF16 |
| 权重格式 | Safetensors，共 4 个分片 |
| 权重数据大小 | 约 16.38 GB |
快速使用
安装依赖：
pip install "transformers>=4.55.4,<5" accelerate torch
加载模型并生成文本：
from transformers import AutoModelForCausalLM, AutoTokenizer
model_id = "SII-LancelotXie/Socio-Foundation-8B"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
model_id,
torch_dtype="auto",
device_map="auto",
)
messages = [{"role": "user", "content": "请介绍一下你自己。"}]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True,
enable_thinking=False,
)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=256, use_cache=True)
response = outputs[0, inputs["input_ids"].shape[1]:]
print(tokenizer.decode(response, skip_special_tokens=True))
说明
本仓库包含完整模型权重，可直接加载，无需另行合并适配器。训练细节与评测结果请参阅上述论文。
许可证
Apache-2.0。
Configuration
- Architecture
- Qwen3ForCausalLM
- Context length (tokens)
- 40,960
- 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
- SII-LancelotXie/Socio-Foundation-8B
- Publisher
- Wenxuan Xie(SII)
- Task
- Text generation
- Modality
- Text
- Library
- transformers
- Parameters
- 8.2B parameters
- Languages
- Not stated by the source
- Revision
- 537c97d78ee8924c12492a406ff667872c750a80
- First published
- 2026-10-06
- Last updated
- 2026-10-09
Files and Weights
16 files, 16.4 GB in total. The weights are 4 files totalling 16.4 GB in safetensors.
Weights4 files · 16.4 GB
Configuration5 files · 36.0 KB
Tokenizer4 files · 15.9 MB
Documentation1 file · 2.0 KB
Other1 file · 4.2 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00004.safetensors | Weights | 4.9 GB | 431d9bac1f5d |
| model-00002-of-00004.safetensors | Weights | 4.9 GB | 6b220aa0cd02 |
| model-00003-of-00004.safetensors | Weights | 5.0 GB | a3fa1165c5a8 |
| model-00004-of-00004.safetensors | Weights | 1.6 GB | 85d5d0324713 |
| added_tokens.json | Configuration | 707 B | — |
| config.json | Configuration | 1.5 KB | — |
| generation_config.json | Configuration | 214 B | — |
| model.safetensors.index.json | Configuration | 32.9 KB | — |
| special_tokens_map.json | Configuration | 613 B | — |
| README.md | Documentation | 2.0 KB | — |
| chat_template.jinja | Other | 4.2 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| merges.txt | Tokenizer | 1.7 MB | — |
| tokenizer.json | Tokenizer | 11.4 MB | aeb13307a71a |
| tokenizer_config.json | Tokenizer | 5.4 KB | — |
| vocab.json | Tokenizer | 2.8 MB | — |
License and Download
- License
- apache-2.0
- Access
- Open weights, no gate
- Download size
- 16.4 GB
Download from Wenxuan Xie(SII)
Released by Wenxuan Xie(SII) through its official repository on Hugging Face. Read the license.
Built From
- Described by arXiv:2610.08967
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.
Questions About Socio-Foundation-8B
How much GPU memory does Socio-Foundation-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 Socio-Foundation-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 Socio-Foundation-8B commercially?
Yes. Socio-Foundation-8B 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 Socio-Foundation-8B's context length?
40,960 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…
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…
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.…