ScitiX offers highly flexible GPU instance pricing, providing you with complete control over configuration and costs to meet your specific needs.
Please note that storage and network costs are not included.
The hourly cost of each instance is determined by the type and quantity of GPUs selected.
$
2.5
- vCPU 22
- VRAM 80GB
- RAM 220GB
$
5.5
- vCPU 31
- VRAM 180GB
- RAM 220GB
$
3.5
- vCPU 22
- VRAM 141GB
- RAM 220GB
$
1.49
- vCPU 15
- VRAM 80GB
- RAM 190GB
| Base Model | $ 1M tokens |
| openai/gpt-oss-120b | $ 0.4 |
| Qwen/Qwen3-235B-A22B-Thinking-2507 | $ 0.4 |
| deepseek-ai/DeepSeek-V3.1-Terminus | $ 0.4 |
| Qwen/Qwen3-32B | $ 0.4 Input, $ 0.32 Output |
| deepseek-ai/DeepSeek-R1 | $ 0.45 Input, $ 1.8 Output |
| Qwen/Qwen3-235B-A22B-Instruct-2507 | $ 0.4 |
| deepseek-ai/DeepSeek-V3.1 | $ 0.04 |
| google/gemma-3-27b-it | $ 0.14 |
| openai/gpt-oss-20b | $ 0.07 |
| Qwen/Qwen2.5-72B-Instruct | $ 0.46 |
| Qwen/Qwen2.5-VL-72B-Instruct | $ 0.46 |
With multi-dimensional complexity abstracted away, you can launch environments instantly — complete with built-in software adaptation, mirror management, and remote S3 access. Our self-developed AI platform integrates with familiar tools like VS Code and JupyterLab.
With ScitiX, there are no hidden costs or complex tiers. You see exactly what you pay for — compute, storage, and bandwidth — in real time. Every GPU-hour and byte transferred is fully trackable, so you can scale confidently without worrying about unexpected charges.
Pay only for what you use. ScitiX bills by actual GPU runtime, down to the minute, making it ideal for dynamic or burst workloads. Pause, resume, or autoscale your jobs anytime while keeping full visibility into cost and performance efficiency.
Save more as you scale. ScitiX offers tailored discounts for reserved usage, long-term commitments, and high-volume workloads. Mix on-demand and reserved instances to find your perfect balance between flexibility and cost efficiency — all with transparent, predictable pricing.
| CPU Model | RAM per vCPU | Cost per vCPU |
|---|---|---|
| Intel® Xeon® Gold 6430 | 16 | $0.05/hour |
| Service | Unit Price | |
|---|---|---|
| Bandwidth | $1 / Mbps / day | $11 / Mbps / month |
| Eip | $4 / per EIP / month | |
| Product | Performance | IO | Price (per GiB/hour) | Price (per TB/day) |
|---|---|---|---|---|
| OSS Standard | standard | - | $0.00007 | $0.00007 |
| xStor Performance | High Performance | Read 1 GB/s/TiB | $0.00014 | $3.416 |
| Write 500MB/s/TiB | ||||
| xStor Capacity | Large Capacity | Read/Write 5MB/s/TiB | $0.00007 | $1.708 |
| Max FS bandwidth 20GB/s |
Self-service
With self-service ordering and a comprehensive user guide to walk you through every step, the docs have you covered whether you're prototyping or scaling.
Open Source
Our infrastructure is built in the open - transparent, verifiable, and always
evolving with the community.
Explore our GitHub to see what we're building, contribute your ideas, or just star
the repo to stay connected.
Contact Us
