RTX PRO 4000 Blackwell 24GB, what it runs and what it costs to rent
Specs checked · Rental prices from · Aliteq Team · How we compute this
Some links on this page are referral links: we may earn a commission, at no cost to you.
The short answer
The NVIDIA RTX PRO 4000 Blackwell has 24 GB of VRAM and 672 GB/s of memory bandwidth (2025). At Q4_K_M quality or better with an 8K context it runs 38 of the open models we track; the largest that fits with room to spare is Qwen3.5 35B-A3B (MoE). The cheapest rental right now is $0.133/hr (Vast.ai, interruptible).
- VRAM
- 24 GB
- Bandwidth
- 672 GB/s
- Board power
- 145 W
- Launched
- 2025
Source: NVIDIA specification · bandwidth not yet confirmed by a second source, so we don't estimate speed from it
Rent the RTX PRO 4000 Blackwell or buy one
| Provider | Type | From, per hour | Rent |
|---|---|---|---|
| Vast.ai | Spot (can be interrupted) | $0.133/hr | Vast.ai |
| Vast.ai | On demand | $0.190/hr | Vast.ai |
| Lium | On demand | $0.270/hr | Lium |
| Runpod | On demand | $0.570/hr | Runpod |
Lowest price per provider in our hourly snapshot (). Rent buttons are referral links. Every card side by side: cloud GPU prices.
Before you rent the RTX PRO 4000 Blackwell: Runpod vs Vast.ai vs Lambda · is Vast.ai safe · is Runpod worth it · rent or buy · every AI GPU's buy and rent price.
RTX PRO 4000 Blackwell rental price history
The cheapest RTX PRO 4000 Blackwell offer we saw each day on the providers we track, since Aug 20. Renting it on demand costs $0.190/hr now; the lowest since Nov 2 was $0.162/hr on Nov 4 (Vast.ai), so today is 18% above that.
- On demand now
- $0.190/hr
- Lowest on demand since Nov 2
- $0.162/hr · Nov 4
- Cheapest offer now
- $0.133/hr
- Lowest offer since Aug 20
- $0.100/hr · Nov 16
- Cheapest offer, incl. interruptible
- Cheapest on demand (from Nov 2)
Daily minimum of our hourly readings of Vast.ai, Runpod and Lium. Interruptible (spot) offers can be taken back mid-job. Before Nov 2 Vast mixed spot into on-demand and Runpod listed Community Cloud rates, so the on-demand line starts then. Every number is in our open daily price file.
Every day as a table
| Day | Cheapest offer | Cheapest on demand |
|---|---|---|
| Nov 18 (now) | $0.133/hr · Vast.ai | $0.190/hr · Vast.ai |
| Nov 17 | $0.100/hr · Vast.ai | $0.174/hr · Vast.ai |
| Nov 16 | $0.100/hr · Vast.ai | $0.164/hr · Vast.ai |
| Nov 15 | $0.107/hr · Vast.ai | $0.175/hr · Vast.ai |
| Nov 14 | $0.107/hr · Vast.ai | $0.189/hr · Vast.ai |
| Nov 13 | $0.107/hr · Vast.ai | $0.204/hr · Vast.ai |
| Nov 12 | $0.107/hr · Vast.ai | $0.189/hr · Vast.ai |
| Nov 11 | $0.107/hr · Vast.ai | $0.189/hr · Vast.ai |
| Nov 10 | $0.107/hr · Vast.ai | $0.189/hr · Vast.ai |
| Nov 9 | $0.107/hr · Vast.ai | $0.189/hr · Vast.ai |
| Nov 8 | $0.107/hr · Vast.ai | $0.175/hr · Vast.ai |
| Nov 7 | $0.107/hr · Vast.ai | $0.176/hr · Vast.ai |
| Nov 6 | $0.133/hr · Vast.ai | $0.164/hr · Vast.ai |
| Nov 5 | $0.107/hr · Vast.ai | $0.163/hr · Vast.ai |
| Nov 4 | $0.107/hr · Vast.ai | $0.162/hr · Vast.ai |
| Nov 3 | $0.147/hr · Vast.ai | $0.227/hr · Vast.ai |
| Nov 2 | $0.107/hr · Vast.ai | $0.162/hr · Vast.ai |
| Nov 1 | $0.107/hr · Vast.ai | — |
| Oct 31 | $0.227/hr · Vast.ai | — |
| Oct 30 | $0.228/hr · Vast.ai | — |
| Oct 29 | $0.228/hr · Vast.ai | — |
| Oct 28 | $0.241/hr · Vast.ai | — |
| Oct 27 | $0.229/hr · Vast.ai | — |
| Oct 26 | $0.241/hr · Vast.ai | — |
| Oct 25 | $0.242/hr · Vast.ai | — |
| Oct 24 | $0.229/hr · Vast.ai | — |
| Oct 23 | $0.256/hr · Vast.ai | — |
| Oct 22 | $0.204/hr · Vast.ai | — |
| Oct 21 | $0.239/hr · Vast.ai | — |
| Oct 20 | $0.204/hr · Vast.ai | — |
| Oct 19 | $0.226/hr · Vast.ai | — |
| Oct 18 | $0.204/hr · Vast.ai | — |
| Oct 17 | $0.188/hr · Vast.ai | — |
| Oct 16 | $0.200/hr · Vast.ai | — |
| Oct 15 | $0.202/hr · Vast.ai | — |
| Oct 14 | $0.188/hr · Vast.ai | — |
| Oct 13 | $0.188/hr · Vast.ai | — |
| Oct 12 | $0.203/hr · Vast.ai | — |
| Oct 11 | $0.215/hr · Vast.ai | — |
| Oct 10 | $0.204/hr · Vast.ai | — |
| Oct 9 | $0.174/hr · Vast.ai | — |
| Oct 8 | $0.190/hr · Vast.ai | — |
| Oct 7 | $0.161/hr · Vast.ai | — |
| Oct 6 | $0.161/hr · Vast.ai | — |
| Oct 5 | $0.187/hr · Vast.ai | — |
| Oct 4 | $0.202/hr · Vast.ai | — |
| Oct 3 | $0.174/hr · Vast.ai | — |
| Oct 2 | $0.202/hr · Vast.ai | — |
| Oct 1 | $0.202/hr · Vast.ai | — |
| Sep 30 | $0.164/hr · Vast.ai | — |
| Sep 29 | $0.202/hr · Vast.ai | — |
| Sep 28 | $0.163/hr · Vast.ai | — |
| Sep 27 | $0.164/hr · Vast.ai | — |
| Sep 26 | $0.183/hr · Vast.ai | — |
| Sep 25 | $0.202/hr · Vast.ai | — |
| Sep 24 | $0.136/hr · Vast.ai | — |
| Sep 23 | $0.188/hr · Vast.ai | — |
| Sep 22 | $0.190/hr · Vast.ai | — |
| Sep 21 | $0.145/hr · Vast.ai | — |
| Sep 20 | $0.197/hr · Vast.ai | — |
| Sep 19 | $0.197/hr · Vast.ai | — |
| Sep 18 | $0.201/hr · Vast.ai | — |
| Sep 17 | $0.190/hr · Vast.ai | — |
| Sep 16 | $0.201/hr · Vast.ai | — |
| Sep 15 | $0.190/hr · Vast.ai | — |
| Sep 14 | $0.134/hr · Vast.ai | — |
| Sep 13 | $0.188/hr · Vast.ai | — |
| Sep 12 | $0.175/hr · Vast.ai | — |
| Sep 11 | $0.134/hr · Vast.ai | — |
| Sep 10 | $0.134/hr · Vast.ai | — |
| Sep 9 | $0.134/hr · Vast.ai | — |
| Sep 8 | $0.134/hr · Vast.ai | — |
| Sep 7 | $0.134/hr · Vast.ai | — |
| Sep 6 | $0.134/hr · Vast.ai | — |
| Sep 5 | $0.134/hr · Vast.ai | — |
| Sep 4 | $0.134/hr · Vast.ai | — |
| Sep 3 | $0.134/hr · Vast.ai | — |
| Sep 2 | $0.134/hr · Vast.ai | — |
| Sep 1 | $0.134/hr · Vast.ai | — |
| Aug 31 | $0.148/hr · Vast.ai | — |
| Aug 30 | $0.148/hr · Vast.ai | — |
| Aug 29 | $0.148/hr · Vast.ai | — |
| Aug 28 | $0.134/hr · Vast.ai | — |
| Aug 27 | $0.134/hr · Vast.ai | — |
| Aug 26 | $0.202/hr · Vast.ai | — |
| Aug 25 | $0.134/hr · Vast.ai | — |
| Aug 24 | $0.202/hr · Vast.ai | — |
| Aug 23 | $0.202/hr · Vast.ai | — |
| Aug 22 | $0.202/hr · Vast.ai | — |
| Aug 21 | $0.174/hr · Vast.ai | — |
| Aug 20 | $0.149/hr · Vast.ai | — |
Which AI models run on the RTX PRO 4000 Blackwell
Computed by our VRAM engine for an 8K context: the best quality each model fits at in 24 GB. Speeds are upper-bound estimates from memory bandwidth, not benchmarks.
| Model | Best fit | VRAM used | ~tok/s |
|---|---|---|---|
| Qwen3.5 35B-A3B (MoE) | Q4_K_M | 21 GB88% | — |
| Qwen3.6 35B-A3B (MoE) | Q4_K_M | 21 GB88% | — |
| Ornith-1.5 35B-A3B (MoE) | Q4_K_M | 21 GB88% | — |
| DeepSeek R1 Distill 32B | Q5_0 | 24 GB100% · tight | — |
| Qwen2.5 32B Instruct | Q5_0 | 24 GB100% · tight | — |
| Qwen2.5 Coder 32B | Q5_0 | 24 GB100% · tight | — |
| Qwen3 32B | Q5_0 | 24 GB100% · tight | — |
| NVIDIA Nemotron 3 Nano 30B-A3B (MoE) | Q5_K_M | 22 GB90% · tight | — |
| Gemma 4 31B | Q5_K_M | 22 GB92% · tight | — |
| GLM-4.7-Flash (MoE) | Q5_K_M | 22 GB91% · tight | — |
| Qwen3-VL 30B-A3B Instruct (MoE) | Q5_K_M | 22 GB92% · tight | — |
| Qwen3 30B-A3B (MoE) | Q5_K_M | 22 GB91% · tight | — |
Fits only below Q4 quality: Mixtral 8x7B (MoE) (Q3_K_M).
Too big for 24 GB (rent a bigger card instead): what it costs to run Llama 3.3 70B Instruct, what it costs to run DeepSeek R1 Distill 70B, what it costs to run Qwen3-Coder-Next 80B-A3B (MoE), what it costs to run Qwen3-Next 80B-A3B (MoE).
Check a model and context of your own in the VRAM calculator.
AI video on the RTX PRO 4000 Blackwell
Alibaba's Wan 2.2 is an open video model you can run yourself in ComfyUI. With 24 GB the RTX PRO 4000 Blackwell fits:
- Wan 2.2 14B at 640x640 with fp8 weights (24 GB and up, per Runpod's Wan 2.2 guide)
- Wan 2.2 5B, with ComfyUI's native offloading (8 GB and up, per ComfyUI's Wan 2.2 docs)
What each setup needs, the cheapest card to rent and what one clip costs: cloud GPUs for ComfyUI with Flux and Wan 2.2. Against an app or an API: AI video cost per clip, app vs API vs rented GPU. Which open model to run, and the licence catch for US users: the best open source AI video generator. Picking by job instead (YouTube, TikTok, ads, animation)? See the best AI video generator for each job. Rather not run a model yourself? Hosted video models such as Seedance, Veo and Hailuo are compared in Kling AI alternatives, priced per clip.
Cards to compare it with
- RTX A5000 · 24 GB
- RTX 3090 · 24 GB
- AMD Radeon RX 7900 XTX · 24 GB
- RTX 3090 Ti · 24 GB
- RTX 4090 · 24 GB
- AMD Radeon RX 7900 XT · 20 GB
Every card by VRAM tier: all GPUs for gaming and local AI.