# Llama 3.3 70B Instruct

> OnPrem World · https://onpremworld.com/en/models/llama-3.3-70b · Last updated: 2026-10-16 · CC BY 4.0

A dense 70B model: at 4-bit it needs a 48 GB card or a 64 GB unified-memory machine, and because every token reads all 70B weights, it is slow on low-bandwidth hardware.

|  |  |
|---|---|
| Developer | Meta |
| Parameters | 70B |
| Native context | 131,072 tokens |
| Licence | Llama 3.3 Community License |
| Commercial use | Yes |
| KV cache | All 80 layers (standard attention) |
| Released | 2024-12 |

```
ollama run llama3.3:70b
```

## Memory needed

Weights + KV cache + runtime overhead, in GiB. Lower quantization = less memory, slightly lower quality.

| Quantization | 4K | 8K | 32K |
|---|---|---|---|
| Q4_K_M | 43.2 GiB | 44.5 GiB | 52.0 GiB |
| Q5_K_M | 50.4 GiB | 51.7 GiB | 59.2 GiB |
| Q6_K | 58.1 GiB | 59.4 GiB | 66.9 GiB |
| Q8_0 | 74.5 GiB | 75.7 GiB | 83.2 GiB |
| FP16 | 138.7 GiB | 139.9 GiB | 147.4 GiB |

## Which hardware runs it (8K)

| Hardware | Memory | Best quant | tok/s (est.) |
|---|---|---|---|
| DGX Station (GB300) | 252 GB | FP16 | 33 |
| RTX PRO 5000 48GB Blackwell | 48 GB | Q4_K_M | 21 |
| RTX PRO 6000 Blackwell Workstation Edition | 96 GB | Q8_0 | 16 |
| RTX PRO 6000 Blackwell Max-Q Workstation Edition | 96 GB | Q8_0 | 16 |
| RTX PRO 5000 72GB Blackwell | 72 GB | Q6_K | 16 |
| RTX 6000 Ada Generation | 48 GB | Q4_K_M | 15 |
| RTX A6000 | 48 GB | Q4_K_M | 12 |
| M5 Ultra | 512 GB | FP16 | 5.5 |
| M5 Max (18-core CPU / 40-core GPU) | 128 GB | Q8_0 | 5.3 |
| M4 Max (16-core CPU / 40-core GPU) | 128 GB | Q8_0 | 4.7 |
| M5 Pro | 64 GB | Q4_K_M | 4.7 |
| M4 Pro | 64 GB | Q4_K_M | 4.1 |
| M3 Ultra | 512 GB | FP16 | 3.8 |
| RTX Spark (N1X) laptops & compact desktops | 128 GB | Q8_0 | 2.4 |
| DGX Spark | 128 GB | Q8_0 | 2.2 |
| Ryzen AI Max+ 395 (Strix Halo) systems | 128 GB | Q8_0 | 2.1 |
| Ryzen AI Max+ PRO 495 (Gorgon Halo) systems | 192 GB | FP16 | 1.2 |
| GeForce RTX 3060 (12 GB) | 12 GB | Doesn't fit | – |
| GeForce RTX 3090 | 24 GB | Doesn't fit | – |
| GeForce RTX 4060 Ti 16GB | 16 GB | Doesn't fit | – |
| GeForce RTX 4080 SUPER | 16 GB | Doesn't fit | – |
| GeForce RTX 4090 | 24 GB | Doesn't fit | – |
| GeForce RTX 5060 Ti 16GB | 16 GB | Doesn't fit | – |
| GeForce RTX 5070 | 12 GB | Doesn't fit | – |
| GeForce RTX 5070 Ti | 16 GB | Doesn't fit | – |
| GeForce RTX 5080 | 16 GB | Doesn't fit | – |
| GeForce RTX 5090 | 32 GB | Doesn't fit | – |
| Radeon RX 7900 XT | 20 GB | Doesn't fit | – |
| Radeon RX 7900 XTX | 24 GB | Doesn't fit | – |
| Radeon RX 9060 XT 16GB | 16 GB | Doesn't fit | – |
| Radeon RX 9070 XT | 16 GB | Doesn't fit | – |
| Arc B580 | 12 GB | Doesn't fit | – |
| Radeon AI PRO R9700 | 32 GB | Doesn't fit | – |
| Arc Pro B60 | 24 GB | Doesn't fit | – |
| Arc Pro B70 | 32 GB | Doesn't fit | – |
| M4 Max (14-core CPU / 32-core GPU) | 36 GB | Doesn't fit | – |
| M5 | 32 GB | Doesn't fit | – |
| M5 Max (18-core CPU / 32-core GPU) | 36 GB | Doesn't fit | – |
| M6 (16GB configuration, 153 GB/s) | 16 GB | Doesn't fit | – |
| M6 (24GB/32GB configurations, 170 GB/s) | 32 GB | Doesn't fit | – |
| RTX PRO 4500 Blackwell | 32 GB | Doesn't fit | – |
| RTX PRO 4000 Blackwell | 24 GB | Doesn't fit | – |

## Frequently asked questions

### How much VRAM does Llama 3.3 70B Instruct need?

With an 8K context, about 44.5 GiB at Q4_K_M and 75.7 GiB at Q8_0 (weights + KV cache + runtime overhead). Longer contexts need more memory for the KV cache.

### Can Llama 3.3 70B Instruct run on an GeForce RTX 4090?

Llama 3.3 70B Instruct does not fit on the GeForce RTX 4090 (24 GB), even at Q4_K_M with an 8K context.

### Can I run Llama 3.3 70B Instruct on a Mac?

Yes. Llama 3.3 70B Instruct fits on the M3 Ultra (96 GB) at Q6_K with an 8K context, using about 59.4 GiB, at an estimated 9.1 tokens/s.

### Can Llama 3.3 70B Instruct be used commercially?

Llama 3.3 70B Instruct is released under Llama 3.3 Community License. The licence allows commercial use (read its conditions).

## Sources

- https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/MODEL_CARD.md
- https://github.com/meta-llama/llama-models/blob/main/models/llama3_3/LICENSE
- https://huggingface.co/RedHatAI/Llama-3.3-70B-Instruct/resolve/main/config.json
- https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct
- https://huggingface.co/api/models/meta-llama/Llama-3.3-70B-Instruct?expand[]=safetensors
- https://ollama.com/library/llama3.3/tags