RTX 4090: 24 GB, what it runs and what it costs
Specs checked · Rental prices from 50m ago · Aliteq Team · How we compute this
The short answer
The NVIDIA GeForce RTX 4090 has 24 GB of VRAM and 1008 GB/s of memory bandwidth (2022). At Q4_K_M quality or better with an 8K context it runs 31 of the open models we track; the largest that fits with room to spare is Qwen3.6 35B-A3B (MoE). The cheapest rental right now is $0.170/hr (Novita AI, interruptible).
- VRAM
- 24 GB
- Bandwidth
- 1008 GB/s
- Board power
- 450 W
- Launched
- 2022
Source: NVIDIA specification · bandwidth not yet confirmed by a second source, so we don't estimate speed from it
Buy it or rent it
| Provider | Type | From, per hour | Rent |
|---|---|---|---|
| Novita AI | Spot (can be interrupted) | $0.170/hr | Novita AI |
| Vast.ai | Spot (can be interrupted) | $0.200/hr | Vast.ai |
| Vast.ai | On demand | $0.267/hr | Vast.ai |
| Novita AI | On demand | $0.330/hr | Novita AI |
| Lium | On demand | $0.350/hr | Lium |
| Lium | Spot (can be interrupted) | $0.470/hr | Lium |
Lowest price per provider in our hourly snapshot (50m ago). Rent buttons are referral links. Every card side by side: cloud GPU prices.
Which AI models run on the RTX 4090
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 |
|---|---|---|---|
| Qwen3.6 35B-A3B (MoE) | Q4_K_M | 21 GB88% | — |
| Ornith-1.5 35B-A3B (MoE) | Q4_K_M | 21 GB88% | — |
| Qwen3.5 35B-A3B (MoE) | Q4_K_M | 21 GB88% | — |
| DeepSeek R1 Distill 32B | Q5_0 | 24 GB100% · tight | — |
| Qwen2.5 Coder 32B | Q5_0 | 24 GB100% · tight | — |
| Qwen2.5 32B Instruct | 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 | — |
| Qwen3 30B-A3B (MoE) | Q5_K_M | 22 GB91% · tight | — |
| Qwen3-Coder 30B-A3B (MoE) | Q5_K_M | 22 GB91% · tight | — |
| Qwen3.5 27B | Q6_K | 22 GB94% · 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 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), what it costs to run GLM-4.5-Air (MoE).
Check a model and context of your own in the VRAM calculator.
AI video on the RTX 4090
Alibaba's Wan 2.2 is an open video model you can run yourself in ComfyUI. With 24 GB the RTX 4090 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. Rather not run a model yourself? Hosted video models such as Seedance, Veo and Hailuo are compared in Kling AI alternatives, priced per clip.
Our guides on the RTX 4090
- CompareRTX 5080 or RTX 4090 for local AI? The 4090, because 24GB runs the 27–32B models 16GB can't
- CompareRTX 4090 or RTX 3090 for local AI? Same 24GB, so buy the 3090 unless you paste long prompts all day
- Where can you rent an RTX 4090 cheapest, and what can its 24GB actually run?
- AI video cost per clip: an app, an API, or a rented GPU? (And what's legal in the US)
- Is Runpod worth it? Yes for an easy setup, if you budget for storage. Here's the real bill
- Is Vast.ai safe and worth it? What it really costs, and the hidden fees to check first
- What's the best GPU for local AI right now? The card with enough VRAM for your model, at the lowest price
- Can you fine-tune an LLM on your own GPU, and what does it cost?
- Which cloud GPU should you rent to fine-tune an LLM, and what does a run cost?
Cards to compare it with
- RTX 3090 Ti · 24 GB
- AMD Radeon RX 7900 XTX · 24 GB
- RTX 3090 · 24 GB
- RTX A5000 · 24 GB
- AMD Radeon RX 7900 XT · 20 GB
- RTX 5080 · 16 GB
Every card by VRAM tier: all GPUs for gaming and local AI.