← CanIRunLocal · All hardware · All models
Liquid AI · 0.587B parameters · chat, vision.
Ollama weight files: Q8_0 0.85 GB (d1-omni:600m).
Yes on 85 machines in this list. The smallest is M1 8GB: Q8_0, about 1.8 GB, roughly 64 tokens per second, via Ollama. ollama run d1-omni:600m
ollama run d1-omni:600mollama run d1-omni:600mollama run d1-omni:600mollama run d1-omni:600mvllm serve LiquidAI/d1-omni-600M --max-model-len 8192 --gpu-memory-utilization 0.9| Hardware | Runtime | Quant | Memory | Fit | Speed | Command |
|---|---|---|---|---|---|---|
| M1 8GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~64 tok/s | ollama run d1-omni:600m |
| M2 8GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~94 tok/s | ollama run d1-omni:600m |
| M3 8GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~94 tok/s | ollama run d1-omni:600m |
| RTX 3060 8GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~169 tok/s | ollama run d1-omni:600m |
| RTX 3060 Ti 8GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~316 tok/s | ollama run d1-omni:600m |
| RTX 3070 8GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~316 tok/s | ollama run d1-omni:600m |
| RTX 4060 8GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~192 tok/s | ollama run d1-omni:600m |
| RTX 4060 Ti 8GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~203 tok/s | ollama run d1-omni:600m |
| RX 7600 8GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~203 tok/s | ollama run d1-omni:600m |
| RTX 3080 10GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~536 tok/s | ollama run d1-omni:600m |
| M1 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~64 tok/s | ollama run d1-omni:600m |
| M1 Pro 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~188 tok/s | ollama run d1-omni:600m |
| M2 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~94 tok/s | ollama run d1-omni:600m |
| M2 Pro 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~188 tok/s | ollama run d1-omni:600m |
| M3 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~94 tok/s | ollama run d1-omni:600m |
| M4 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~113 tok/s | ollama run d1-omni:600m |
| RTX 3060 12GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~254 tok/s | ollama run d1-omni:600m |
| RTX 3080 12GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~644 tok/s | ollama run d1-omni:600m |
| RTX 4070 12GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~356 tok/s | ollama run d1-omni:600m |
| RTX 5070 12GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~474 tok/s | ollama run d1-omni:600m |
| M3 Pro 18GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~141 tok/s | ollama run d1-omni:600m |
| RTX 4060 Ti 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~203 tok/s | ollama run d1-omni:600m |
| RTX 4070 Ti Super 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~474 tok/s | ollama run d1-omni:600m |
| RTX 4080 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~506 tok/s | ollama run d1-omni:600m |
| RTX 5080 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~678 tok/s | ollama run d1-omni:600m |
| RX 7800 XT 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~440 tok/s | ollama run d1-omni:600m |
| RX 9070 XT 16GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~452 tok/s | ollama run d1-omni:600m |
| M2 24GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~94 tok/s | ollama run d1-omni:600m |
| M3 24GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~94 tok/s | ollama run d1-omni:600m |
| M4 24GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~113 tok/s | ollama run d1-omni:600m |
| M4 Pro 24GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~257 tok/s | ollama run d1-omni:600m |
| RTX 3090 24GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~661 tok/s | ollama run d1-omni:600m |
| RTX 4090 24GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~712 tok/s | ollama run d1-omni:600m |
| RX 7900 XTX 24GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~678 tok/s | ollama run d1-omni:600m |
| M1 Max (24c GPU) 32GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M1 Max (32c GPU) 32GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M1 Pro 32GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~188 tok/s | ollama run d1-omni:600m |
| M2 Max (30c GPU) 32GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M2 Max (38c GPU) 32GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M2 Pro 32GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~188 tok/s | ollama run d1-omni:600m |
| M4 32GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~113 tok/s | ollama run d1-omni:600m |
| M3 Max (30c GPU) 36GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~282 tok/s | ollama run d1-omni:600m |
| M3 Pro 36GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~141 tok/s | ollama run d1-omni:600m |
| M4 Max (32c GPU) 36GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~386 tok/s | ollama run d1-omni:600m |
| Mac Studio M4 Max (32c GPU) 36GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~386 tok/s | ollama run d1-omni:600m |
| Mac Studio M5 Max (32c GPU) 36GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~433 tok/s | ollama run d1-omni:600m |
| RTX 5090 32GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~1265 tok/s | ollama run d1-omni:600m |
| M3 Max (40c GPU) 48GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M4 Max (40c GPU) 48GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~514 tok/s | ollama run d1-omni:600m |
| M4 Pro 48GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~257 tok/s | ollama run d1-omni:600m |
| Mac Studio M4 Max (40c GPU) 48GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~514 tok/s | ollama run d1-omni:600m |
| Mac Studio M5 Max (40c GPU) 48GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~578 tok/s | ollama run d1-omni:600m |
| M1 Max (32c GPU) 64GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M2 Max (30c GPU) 64GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M2 Max (38c GPU) 64GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M2 Ultra 64GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~753 tok/s | ollama run d1-omni:600m |
| M3 Max (40c GPU) 64GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M4 Max (40c GPU) 64GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~514 tok/s | ollama run d1-omni:600m |
| M4 Pro 64GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~257 tok/s | ollama run d1-omni:600m |
| Mac Studio M4 Max (40c GPU) 64GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~514 tok/s | ollama run d1-omni:600m |
| Mac Studio M5 Max (40c GPU) 64GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~578 tok/s | ollama run d1-omni:600m |
| M2 Max (38c GPU) 96GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M3 Max (30c GPU) 96GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~282 tok/s | ollama run d1-omni:600m |
| Mac Studio M3 Ultra (60c GPU) 96GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~771 tok/s | ollama run d1-omni:600m |
| Mac Studio M3 Ultra (80c GPU) 96GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~771 tok/s | ollama run d1-omni:600m |
| Mac Studio M5 Ultra (64c GPU) 96GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~1129 tok/s | ollama run d1-omni:600m |
| Mac Studio M5 Ultra (80c GPU) 96GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~1129 tok/s | ollama run d1-omni:600m |
| M2 Ultra 128GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~753 tok/s | ollama run d1-omni:600m |
| M3 Max (40c GPU) 128GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~376 tok/s | ollama run d1-omni:600m |
| M4 Max (40c GPU) 128GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~514 tok/s | ollama run d1-omni:600m |
| Mac Studio M4 Max (40c GPU) 128GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~514 tok/s | ollama run d1-omni:600m |
| Mac Studio M5 Max (40c GPU) 128GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~578 tok/s | ollama run d1-omni:600m |
| DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte) | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~241 tok/s | ollama run d1-omni:600m |
| M2 Ultra 192GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~753 tok/s | ollama run d1-omni:600m |
| 2x Mac Studio M5 Max (40c GPU) 128GB | MLX | MLX 8bit | 1.7 GB | fits in fast memory | ~942 tok/s | exo # on every Mac (brew install --cask exo); load LiquidAI/d1-omni-600M-MLX-8bit with Tensor sharding + RDMA |
| Mac Studio M3 Ultra (60c GPU) 256GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~771 tok/s | ollama run d1-omni:600m |
| Mac Studio M3 Ultra (80c GPU) 256GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~771 tok/s | ollama run d1-omni:600m |
| Mac Studio M5 Ultra (64c GPU) 256GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~1129 tok/s | ollama run d1-omni:600m |
| Mac Studio M5 Ultra (80c GPU) 256GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~1129 tok/s | ollama run d1-omni:600m |
| 2x DGX Spark 128GB | vLLM | Official weights | 3.3 GB | fits in fast memory | ~109 tok/s | vllm serve LiquidAI/d1-omni-600M --tensor-parallel-size 2 --distributed-executor-backend ray --max-model-len 8192 |
| 4x Mac Studio M5 Max (40c GPU) 128GB | MLX | MLX 8bit | 1.7 GB | fits in fast memory | ~1480 tok/s | exo # on every Mac (brew install --cask exo); load LiquidAI/d1-omni-600M-MLX-8bit with Tensor sharding + RDMA |
| 2x Mac Studio M5 Ultra (80c GPU) 256GB | MLX | MLX 8bit | 1.7 GB | fits in fast memory | ~1841 tok/s | exo # on every Mac (brew install --cask exo); load LiquidAI/d1-omni-600M-MLX-8bit with Tensor sharding + RDMA |
| Mac Studio M3 Ultra (80c GPU) 512GB | Ollama | Q8_0 | 1.8 GB | fits in fast memory | ~771 tok/s | ollama run d1-omni:600m |
| 4x DGX Spark 128GB | vLLM | Official weights | 3.3 GB | fits in fast memory | ~157 tok/s | vllm serve LiquidAI/d1-omni-600M --tensor-parallel-size 4 --distributed-executor-backend ray --max-model-len 8192 |
| 4x Mac Studio M5 Ultra (80c GPU) 256GB | MLX | MLX 8bit | 1.7 GB | fits in fast memory | ~2893 tok/s | exo # on every Mac (brew install --cask exo); load LiquidAI/d1-omni-600M-MLX-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 2026-10-19.
A project by OnDevice Labs · llms.txt