Model reference · open weights
XCurOS0.1 is an open-weight language model from XCurOS. XCurOS0.1-8B-Instruct (BF16) weighs 15.2 GB; the smallest configuration that runs it is 2× RTX 3060 12 GB.
For self-hosting, compare this build's requirements below with the GPU you plan to rent.
Summary of the XCurOS/XCurOS0.1-8B-Instruct model card, 2026-10-01
What it is
| Released by | XCurOS |
|---|---|
| Released | 2026-02-28 |
| Parameters | 7.6B |
| VRAM | 15.2 GB for the weights |
What it runs on
| Card | Requests at once | Context max | Memory | |
|---|---|---|---|---|
| 8K each | 32K each | |||
| RTX 3060 12 GB … RTX 4060 Ti 16 GB 2 smaller cards | ||||
| RTX 3090 24 GB | 15 | 3 | all 32K | 23.4 GB |
| RTX 4090 24 GB | 15 | 3 | all 32K | 23.4 GB |
| RTX 5090 32 GB | 31 | 7 | all 32K | 31.0 GB |
| L40S 48 GB | 59 | 14 | all 32K | 44.0 GB |
| A100 80 GB | 132 | 33 | all 32K | 78.2 GB |
| H100 80 GB | 122 | 30 | all 32K | 78.1 GB |
| RTX PRO 6000 Blackwell 96 GB | 156 | 39 | all 32K | 93.8 GB |
| DGX Spark (GB10) 128 GB unified | 184 | 46 | all 32K | 107 GB |
| H200 141 GB | 250 | 62 | all 32K | 138 GB |
| B200 180 GB | 329 | 82 | all 32K | 176 GB |
| 2× RTX 3060 12 GB tensor parallel | 13 | 3 | all 32K | 11.6 GB a card |
| 2× RTX 4060 Ti 16 GB tensor parallel | 29 | 7 | all 32K | 15.4 GB a card |
| Requests at once | 8K tokens each | 32K tokens each |
|---|---|---|
| 1 | 16.5 GB | 17.9 GB |
| 5 | 18.4 GB | 25.4 GB |
| 8 | 19.8 GB | 31.1 GB |
| 16 | 23.5 GB | 46.1 GB |
| 32 | 31.1 GB | 76.2 GB |
| 64 | 46.1 GB | 136 GB |
One card, with vLLM's small-card settings.
From the model card
XCurOS-0.1-8B-Instruct is a proprietary large language model developed exclusively by XCurOS.
It delivers advanced natural language understanding, reasoning, and instruction-following capabilities, designed for secure enterprise and on-premise environments.
🔐 Security-First Design Built for operation in isolated and sensitive infrastructures with full data privacy.
🧠 Advanced Language Intelligence High-quality reasoning, coding, and complex instruction execution.
🖥 Agent & System Integration Ready Designed for automation pipelines, OS-level agents, and enterprise workflows.
📄 Document Understanding Capable of analyzing structured and unstructured text across large documents.
🎯 Instruction-Tuned Performance Optimized to follow commands accurately and consistently.
⚡ Efficient Deployment Suitable for local servers and cloud infrastructure with optimized performance.
🌐 Long-Context Reasoning Handles long conversations, books, and extensive datasets coherently.
🧩 Extensible Architecture Integrates easily into custom applications and secure enterprise systems.
XCurOS-0.1-8B-Instruct architecture provides:
Model Name: XCurOS-0.1-8B-Instruct Organization: XCurOS Version: 0.0.1 License: Proprietary software, all rights reserved Instruction Tuned: True Secure OS Ready: True Intended Audience: Enterprises, research labs, private users
This model is proprietary software owned by XCurOS.
The model may be used for personal and research purposes only when deployed on a private server or local infrastructure.
Personal use includes:
Commercial use is NOT permitted without an official license.
Any commercial activity including but not limited to:
Requires a separate commercial license from the XCurOS Team. For commercial licensing inquiries, please contact the XCurOS Team.
Unauthorized redistribution, resale, or commercial usage of the model is strictly prohibited without written permission from XCurOS.
Maintained internally by XCurOS.
For licensing, collaboration, or support, contact the XCurOS Team.
Author: XCurOS-Team (f13b696767b224479d7a06bffae9a0b62e38e2e2)
Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.