NVIDIA RTX 4070 GPU: Specs, VRAM, Price & Benchmarks (2026)
The RTX 4070 is a real GPU (launched 2023), but no provider on Aquanode is listing it for rental right now, so there is no live hourly price to quote. Availability changes as providers add and retire hardware; the specs, VRAM and price context below still apply.
How much VRAM does the RTX 4070 have?
The RTX 4070 has 12GB GDDR6X, with 504 GB/s of peak memory bandwidth.
RTX 4070 VRAM calculator: check which models fit in its memory at each precision.
All models that fit in 12 GB: the open models whose weights and overhead fit, at native, FP8 and INT4 precision.
What can the RTX 4070 run?
Popular open models from small to frontier scale, with the memory each needs and how many RTX 4070 cards (12GB GDDR6X each) that takes.
| Model | As published | FP8 | INT4 |
|---|---|---|---|
| Qwen/Qwen3-8B 8.2B | BF16: ~18.3 GB, 2 GPUs | FP8: ~9.2 GB, 1 GPU | INT4: ~4.6 GB, 1 GPU |
| Qwen/Qwen2.5-14B-Instruct 14.8B | BF16: ~33 GB, 3 GPUs | FP8: ~16.5 GB, 2 GPUs | INT4: ~8.3 GB, 1 GPU |
| Qwen/Qwen3-32B 32.8B | BF16: ~73.2 GB, 7 GPUs | FP8: ~36.6 GB, 4 GPUs | INT4: ~18.3 GB, 2 GPUs |
| Qwen/Qwen-72B 72.3B | BF16: ~162 GB, 14 GPUs | FP8: ~80.8 GB, 7 GPUs | INT4: ~40.4 GB, 4 GPUs |
| MiniMaxAI/MiniMax-M2.7 228.7B | FP8: ~256 GB, 22 GPUs | – | INT4: ~128 GB, 11 GPUs |
| deepseek-ai/DeepSeek-R1 684.5B | FP8: ~765 GB, 64 GPUs | – | INT4: ~383 GB, 32 GPUs |
Estimates: weights at the stated precision plus a flat 20% for KV cache and overhead, at a moderate context length. A dash means the precision is not offered for that model (it is already published at that size). INT4 needs a published quantized checkpoint. Open any model for a per-GPU breakdown, or use the RTX 4070 VRAM calculator.
RTX 4070 specs
| Architecture | launched 2023 |
| VRAM | 12GB GDDR6X |
| Memory bandwidth | 504 GB/s |
| FP16 / BF16 tensor throughput | 116.6 TFLOPS (peak, dense) |
| FP8 tensor throughput | 233.2 TFLOPS (peak, dense) |
| TDP | 200W |
| Form factor | PCIe, 2-slot |
Specs sourced from the vendor's public product page. See the source.
GPU Glossary: What is VRAM?, Tensor Cores, CUDA Cores, TFLOPS
Good for
A 200W card with real FP8 tensor cores. The cheapest FP8-capable hardware here after the 4060 Ti class, and enough for quantized 7-8B serving at low cost per hour.
Not good for
12GB and 504 GB/s make this a small-model-only card, and 116.6 dense FP16 TFLOPS is roughly a quarter of an RTX 5080's. Nothing above ~8B fits without quantizing. No ECC, no NVLink.
RTX 4070 price: what does it cost?
Buying. $599 (launch MSRP).
No provider on Aquanode currently lists the RTX 4070 for rental, so there is no live per-GPU-hour rate to show.
RTX 4070 price by region
Related guides
Other models in the same generation. The full list is in the GPU index.
Get notified when the price drops
GPU supply moves hourly. Tell us what you're waiting for and we'll email you when a matching offer appears across any provider we track.
