← CanIRunLocal · All hardware · All models
Meta · 10.7B parameters · vision.
Ollama weight files: Q4_K_M 7.8 GB (llama3.2-vision:11b); Q8_0 12 GB (llama3.2-vision:11b-instruct-q8_0).
Yes on 59 machines in this list. The smallest is RTX 3060 Ti 8GB: Q4_K_M, about 8.9 GB, roughly 18 tokens per second, via Ollama. ollama run llama3.2-vision:11b
ollama run llama3.2-vision:11b-instruct-q8_0ollama run llama3.2-vision:11b-instruct-q8_0ollama run llama3.2-vision:11b-instruct-q8_0ollama run llama3.2-vision:11b-instruct-q8_0vllm serve meta-llama/Llama-3.2-11B-Vision-Instruct --max-model-len 8192 --gpu-memory-utilization 0.9| Hardware | Runtime | Quant | Memory | Fit | Speed | Command |
|---|---|---|---|---|---|---|
| RTX 3060 Ti 8GB | Ollama | Q4_K_M | 8.9 GB | partial offload (slower) | ~18 tok/s | ollama run llama3.2-vision:11b |
| RTX 3070 8GB | Ollama | Q4_K_M | 8.9 GB | partial offload (slower) | ~18 tok/s | ollama run llama3.2-vision:11b |
| RTX 4060 8GB | Ollama | Q4_K_M | 8.9 GB | partial offload (slower) | ~14 tok/s | ollama run llama3.2-vision:11b |
| RX 7600 8GB | Ollama | Q4_K_M | 8.9 GB | partial offload (slower) | ~15 tok/s | ollama run llama3.2-vision:11b |
| RTX 3080 10GB | Ollama | Q4_K_M | 8.9 GB | fits in fast memory | ~58 tok/s | ollama run llama3.2-vision:11b |
| M1 16GB | Ollama | Q4_K_M | 8.9 GB | fits in fast memory | ~7 tok/s | ollama run llama3.2-vision:11b |
| M2 16GB | Ollama | Q4_K_M | 8.9 GB | fits in fast memory | ~10 tok/s | ollama run llama3.2-vision:11b |
| M3 16GB | Ollama | Q4_K_M | 8.9 GB | fits in fast memory | ~10 tok/s | ollama run llama3.2-vision:11b |
| M4 16GB | Ollama | Q4_K_M | 8.9 GB | fits in fast memory | ~12 tok/s | ollama run llama3.2-vision:11b |
| RTX 3060 12GB | Ollama | Q4_K_M | 8.9 GB | fits in fast memory | ~28 tok/s | ollama run llama3.2-vision:11b |
| RTX 4070 12GB | Ollama | Q4_K_M | 8.9 GB | fits in fast memory | ~39 tok/s | ollama run llama3.2-vision:11b |
| RTX 5070 12GB | Ollama | Q4_K_M | 8.9 GB | fits in fast memory | ~52 tok/s | ollama run llama3.2-vision:11b |
| RTX 4060 Ti 16GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~14 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| RTX 4070 Ti Super 16GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~34 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| RTX 4080 16GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~36 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| RTX 5080 16GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~48 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| RX 7800 XT 16GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~31 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| RX 9070 XT 16GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~32 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M2 24GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~7 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M3 24GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~7 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M4 24GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~8 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| RTX 3090 24GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~47 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| RTX 4090 24GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~50 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| RX 7900 XTX 24GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~48 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M1 Pro 32GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~13 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M2 Pro 32GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~13 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M4 32GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~8 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M3 Pro 36GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~10 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M4 Max (32c GPU) 36GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~27 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M5 Max (32c GPU) 36GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~31 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| RTX 5090 32GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~90 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M4 Pro 48GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~18 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M4 Max (40c GPU) 48GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~36 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M5 Max (40c GPU) 48GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~41 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M1 Max 64GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~27 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M2 Max 64GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~27 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M3 Max 64GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~27 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M4 Max 64GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~36 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M4 Max (40c GPU) 64GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~36 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M5 Max (40c GPU) 64GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~41 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| DGX Spark 64GB (ASUS, Dell, HP, MSI, Acer, Gigabyte) | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~17 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M3 Ultra 96GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~55 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M5 Ultra 96GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~80 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M3 Max 128GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~27 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M4 Max 128GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~36 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M4 Max (40c GPU) 128GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~36 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M5 Max (40c GPU) 128GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~41 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte) | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~17 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| M2 Ultra 192GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~53 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| 2x Mac Studio M5 Max (40c GPU) 128GB | MLX | MLX 8bit | 12.5 GB | fits in fast memory | ~61 tok/s | exo # on every Mac (brew install --cask exo); load mlx-community/Llama-3.2-11B-Vision-Instruct-8bit with Tensor sharding + RDMA |
| Mac Studio M3 Ultra 256GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~55 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M5 Ultra (64c GPU) 256GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~80 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| Mac Studio M5 Ultra (80c GPU) 256GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~80 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| 2x DGX Spark 128GB | vLLM | Official weights | 22.5 GB | fits in fast memory | ~12 tok/s | vllm serve meta-llama/Llama-3.2-11B-Vision-Instruct --tensor-parallel-size 2 --distributed-executor-backend ray --max-model-len 8192 |
| 4x Mac Studio M5 Max (40c GPU) 128GB | MLX | MLX 8bit | 12.5 GB | fits in fast memory | ~95 tok/s | exo # on every Mac (brew install --cask exo); load mlx-community/Llama-3.2-11B-Vision-Instruct-8bit with Tensor sharding + RDMA |
| 2x Mac Studio M5 Ultra (80c GPU) 256GB | MLX | MLX 8bit | 12.5 GB | fits in fast memory | ~119 tok/s | exo # on every Mac (brew install --cask exo); load mlx-community/Llama-3.2-11B-Vision-Instruct-8bit with Tensor sharding + RDMA |
| Mac Studio M3 Ultra (80c GPU) 512GB | Ollama | Q8_0 | 13.1 GB | fits in fast memory | ~55 tok/s | ollama run llama3.2-vision:11b-instruct-q8_0 |
| 4x DGX Spark 128GB | vLLM | Official weights | 22.5 GB | fits in fast memory | ~17 tok/s | vllm serve meta-llama/Llama-3.2-11B-Vision-Instruct --tensor-parallel-size 4 --distributed-executor-backend ray --max-model-len 8192 |
| 4x Mac Studio M5 Ultra (80c GPU) 256GB | MLX | MLX 8bit | 12.5 GB | fits in fast memory | ~186 tok/s | exo # on every Mac (brew install --cask exo); load mlx-community/Llama-3.2-11B-Vision-Instruct-8bit with Tensor sharding + RDMA |
Estimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-16.
A project by OnDevice Labs · llms.txt