Best GPU for DeepSeek R1: VRAM Requirements
DeepSeek R1 needs about 1,476.8 GB of GPU memory in BF16 and 369.7 GB at 4 bits. See which GPUs fit it for inference and fine-tuning.
DeepSeek R1 needs about 1,476.8 GB of VRAM in FP16, 738.7 GB in FP8 and 369.7 GB at 4 bits for inference with an 8,192-token context. In FP16 it fits on 8 × NVIDIA B300 288GB; a 4-bit build fits on 2 × NVIDIA B300 288GB. Its weights are published in FP8, so the FP16 figure is for reference.
DeepSeek R1 is a mixture-of-experts language model with 671B total and 37B active parameters per token from DeepSeek, released in January 2025, with a 131,072-token context window and published under the MIT. Its weights alone take 1,342 GB in BF16 and 335.5 GB at 4 bits per parameter; the tables below add the KV cache and runtime memory, then list the GPUs that hold each setup.
Recommended GPUs for DeepSeek R1
Best for inference
4 GPUs × NVIDIA B300 288GB
FP8 or INT8 weights, 738.7 GB at 8,192 tokens.
Best for fine-tuning
8 GPUs × NVIDIA B300 288GB
LoRA on BF16 weights, 1,744.6 GB.
Budget pick
8 GPUs × NVIDIA RTX PRO 6000 Blackwell 96GB
Smallest card class that holds a 4-bit build, 369.7 GB.
DeepSeek R1 VRAM requirements by precision
One request with an 8,192-token context, KV cache in BF16, plus a 10 % margin for activations and runtime buffers.
| Precision | Weights | KV cache | Total for inference |
|---|---|---|---|
| BF16 / FP16 | 1,342 GB | 0.6 GB | 1,476.8 GB |
| FP8 (published weights) | 671 GB | 0.6 GB | 738.7 GB |
| INT4 | 335.5 GB | 0.6 GB | 369.7 GB |
DeepSeek R1 VRAM by task
Fine-tuning rows assume a short sequence, batch 1 and gradient checkpointing; longer sequences add activation memory.
| Task | GPU memory |
|---|---|
| Inference at the full 131,072-token context (FP8) | 748.2 GB |
| QLoRA (4-bit base) | 436.2 GB |
| LoRA (BF16 base) | 1,744.6 GB |
| Full fine-tuning (BF16, AdamW) | 11,809.6 GB |
GPUs that fit DeepSeek R1
Number of cards whose combined memory holds each setup, assuming the model is split across them with tensor or pipeline parallelism. A dash means more than eight cards.
| GPU | Memory | BF16 inference (1,476.8 GB) | FP8 inference (738.7 GB) | INT4 inference (369.7 GB) | QLoRA (436.2 GB) | LoRA (1,744.6 GB) |
|---|---|---|---|---|---|---|
| NVIDIA L40S 48GB | 48 GB | — | — | 8 GPUs | — | — |
| NVIDIA Quadro RTX 8000 48GB | 48 GB | — | — | 8 GPUs | — | — |
| NVIDIA RTX 6000 Ada 48GB | 48 GB | — | — | 8 GPUs | — | — |
| NVIDIA RTX A6000 48GB | 48 GB | — | — | 8 GPUs | — | — |
| NVIDIA A100 PCIe 80GB | 80 GB | — | — | 8 GPUs | 8 GPUs | — |
| NVIDIA A100 SXM 80GB | 80 GB | — | — | 8 GPUs | 8 GPUs | — |
| NVIDIA H100 PCIe 80GB | 80 GB | — | — | 8 GPUs | 8 GPUs | — |
| NVIDIA H100 SXM 80GB | 80 GB | — | — | 8 GPUs | 8 GPUs | — |
| NVIDIA RTX PRO 6000 Blackwell 96GB | 96 GB | — | 8 GPUs | 4 GPUs | 8 GPUs | — |
| NVIDIA H200 141GB | 141 GB | — | 8 GPUs | 4 GPUs | 4 GPUs | — |
| NVIDIA B200 180GB | 180 GB | — | 8 GPUs | 4 GPUs | 4 GPUs | — |
| NVIDIA B300 288GB | 288 GB | 8 GPUs | 4 GPUs | 2 GPUs | 2 GPUs | 8 GPUs |
| NVIDIA GB300 NVL | 288 GB | 8 GPUs | 4 GPUs | 2 GPUs | 2 GPUs | 8 GPUs |
What is DeepSeek R1?
DeepSeek R1 is a reasoning model trained with large-scale reinforcement learning on the DeepSeek V3 base: a 671-billion-parameter mixture-of-experts model that activates 37 billion parameters per token. Its weights are published in FP8, so FP8 is its full-quality format.
- Developer
- DeepSeek
- Released
- January 2025
- Parameters
- 671B total, 37B active per token
- Context window
- 131,072 tokens
- License
- MIT
Multi-head latent attention caches one compressed 576-value vector per layer and token instead of separate keys and values; the estimate counts it as a KV width of 288. The May 2025 update, DeepSeek R1 0528, keeps the same architecture and size, so these figures apply to it too.
Source: DeepSeek R1 model card (as of November 17, 2026).
Frequently asked questions
- How much VRAM does DeepSeek R1 need?
- For inference with an 8,192-token context and one request, DeepSeek R1 needs about 1,476.8 GB in BF16, 738.7 GB in FP8, 369.7 GB in INT4, including a 10 % runtime margin. Real usage depends on the inference engine and its settings.
- Can DeepSeek R1 run on a single GPU?
- Not among the GPUs listed here: even a 4-bit build needs about 369.7 GB, so the model is split across several GPUs with tensor or pipeline parallelism.
- How much GPU memory does fine-tuning DeepSeek R1 take?
- QLoRA (4-bit base): about 436.2 GB; LoRA (BF16 base): about 1,744.6 GB; Full fine-tuning (BF16, AdamW): about 11,809.6 GB. Longer training sequences and larger batches add activation memory on top.
- What is the best GPU for DeepSeek R1?
- Best for inference: 4 GPUs × NVIDIA B300 288GB (FP8 or INT8 weights, 738.7 GB at 8,192 tokens). Best for fine-tuning: 8 GPUs × NVIDIA B300 288GB (LoRA on BF16 weights, 1,744.6 GB). Budget pick: 8 GPUs × NVIDIA RTX PRO 6000 Blackwell 96GB (Smallest card class that holds a 4-bit build, 369.7 GB). Check live offers before booking, as availability changes.
- Can DeepSeek R1 run on an RTX 4090?
- No: even a 4-bit build needs about 369.7 GB, more than eight RTX 4090 cards hold (192 GB). It needs data center GPUs with more memory per card.
- Can DeepSeek R1 run on an RTX 5090?
- No: even a 4-bit build needs about 369.7 GB, more than eight RTX 5090 cards hold (256 GB). It needs data center GPUs with more memory per card.
Estimates use the formulas of the LLM VRAM calculator (decimal GB). Real usage depends on the inference engine, quantization format and memory fragmentation: test on the target GPU before a long rental.
Explore the live market
Every offer on the live market comes from a reviewed supplier and is sold by EverXCloud Inc.