RTX PRO 4000
The RTX PRO 4000 is a cloud GPU with 24 GB of memory. It is currently available from 2 providers starting at $0.22/hr, with a market median of $0.50/hr across 5 offerings.
Hardware specifications
- Architecture
- Blackwell
- VRAM
- 24 GB
Same across all providers.
Market comparison
All-time price history
Offerings
No offerings found
Try clearing the spot filter or check back later.
LLMs that fit
| Model | Params | 16-bit | 8-bit | 4-bit |
|---|---|---|---|---|
Laguna-S 2.1 Poolside | 118B 8B active | Needs >8 GPUs | Barely fits on 8× | Barely fits on 5× |
gpt-oss-120b OpenAI | 117B 5.1B active | Not released | Not released | Barely fits on 4× |
Llama 3.1 70B Meta | 70B | Barely fits on 7× | Barely fits on 4× | 3× · 18 GB free |
Qwen3.6 35B-A3B Alibaba | 35B 3B active | 5× · 14% spare | 3× · 15.6 GB free | 2× · 14.4 GB free |
Qwen3 32B Alibaba | 32.8B | Barely fits on 4× | 3× · 20.4 GB free | 2× · 18 GB free |
Qwen2.5 14B Alibaba | 14B | 2× · 12.6 GB free | 1× · 4.7 GB free | 1× · 12.6 GB free |
Llama 3.1 8B Meta | 8B | Barely fits on 1× | 1× · 12.6 GB free | 1× · 17.3 GB free |
Mistral 7B Mistral AI | 7B | 1× · 5.8 GB free | 1× · 14 GB free | 1× · 18.1 GB free |
Qwen3 4B Alibaba | 4B | 1× · 12.4 GB free | 1× · 17.6 GB free | 1× · 20.3 GB free |
Estimates: model weights plus 20% for KV cache and runtime overhead, at short context lengths. Long contexts and large batches need more. Rows marked “tight” hold the weights but not that full margin. A ×N figure is the total VRAM across N of these GPUs and makes no claim about interconnect throughput.
Similar GPUs
See more GPUs like this in
Frequently asked questions
How much does the RTX PRO 4000 cost per hour?
The cheapest RTX PRO 4000 offering starts at $0.22/hr, and the market median is $0.50/hr.
Which cloud providers offer the RTX PRO 4000?
No providers currently list the RTX PRO 4000 in our catalog.
How much VRAM does the RTX PRO 4000 have?
The RTX PRO 4000 has 24 GB of VRAM.
What LLMs can run on the RTX PRO 4000?
With 24 GB of VRAM, the RTX PRO 4000 can run 9 of our tracked models in 4-bit quantization, including Laguna-S 2.1, gpt-oss-120b, Qwen3.6 35B-A3B. 7 models fit in 16-bit precision.
What's the biggest LLM I can run on multiple RTX PRO 4000s?
With 8× RTX PRO 4000 (192 GB total), the largest tracked model that fits in 16-bit is Qwen3.6 35B-A3B (35B params), which needs 5× GPUs.