← CanIRunLocal · All hardware · All models
Ornith · 35B parameters (3B active) · chat, reasoning, coding, vision.
Ollama weight files: Default 23 GB (ornith-1.5:35b).
Yes on 44 machines in this list. The smallest is RTX 4060 Ti 16GB: Default, about 24.5 GB, roughly 20 tokens per second, via Ollama. ollama run ornith-1.5:35b
ollama run ornith-1.5:35bollama run ornith-1.5:35bollama run ornith-1.5:35bollama run ornith-1.5:35blms get ornith-ai/Ornith-1.5-35B-A3B-GGUF@Q8_0llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q8_0 -c 8192vllm serve ornith-ai/Ornith-1.5-35B-A3B --max-model-len 8192 --gpu-memory-utilization 0.9| Hardware | Runtime | Quant | Memory | Fit | Speed | Command |
|---|---|---|---|---|---|---|
| RTX 4060 Ti 16GB | Ollama | Default | 24.5 GB | partial offload (slower) | ~20 tok/s | ollama run ornith-1.5:35b |
| RTX 4070 Ti Super 16GB | Ollama | Default | 24.5 GB | partial offload (slower) | ~24 tok/s | ollama run ornith-1.5:35b |
| RTX 4080 16GB | Ollama | Default | 24.5 GB | partial offload (slower) | ~24 tok/s | ollama run ornith-1.5:35b |
| RTX 5080 16GB | Ollama | Default | 24.5 GB | partial offload (slower) | ~25 tok/s | ollama run ornith-1.5:35b |
| RX 7800 XT 16GB | Ollama | Default | 24.5 GB | partial offload (slower) | ~23 tok/s | ollama run ornith-1.5:35b |
| RX 9070 XT 16GB | Ollama | Default | 24.5 GB | partial offload (slower) | ~23 tok/s | ollama run ornith-1.5:35b |
| RTX 3090 24GB | LM Studio | Q4_K_M GGUF | 23.2 GB | fits in fast memory | ~151 tok/s | lms get ornith-ai/Ornith-1.5-35B-A3B-GGUF@Q4_K_M |
| RTX 4090 24GB | LM Studio | Q4_K_M GGUF | 23.2 GB | fits in fast memory | ~163 tok/s | lms get ornith-ai/Ornith-1.5-35B-A3B-GGUF@Q4_K_M |
| RX 7900 XTX 24GB | LM Studio | Q4_K_M GGUF | 23.2 GB | fits in fast memory | ~155 tok/s | lms get ornith-ai/Ornith-1.5-35B-A3B-GGUF@Q4_K_M |
| M1 Pro 32GB | llama.cpp | Q4_K_M GGUF | 23.2 GB | fits in fast memory | ~43 tok/s | llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q4_K_M -c 8192 |
| M2 Pro 32GB | llama.cpp | Q4_K_M GGUF | 23.2 GB | fits in fast memory | ~43 tok/s | llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q4_K_M -c 8192 |
| M4 32GB | llama.cpp | Q4_K_M GGUF | 23.2 GB | fits in fast memory | ~26 tok/s | llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q4_K_M -c 8192 |
| M3 Pro 36GB | Ollama | Default | 24.5 GB | fits in fast memory | ~30 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M4 Max (32c GPU) 36GB | Ollama | Default | 24.5 GB | fits in fast memory | ~83 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M5 Max (32c GPU) 36GB | Ollama | Default | 24.5 GB | fits in fast memory | ~93 tok/s | ollama run ornith-1.5:35b |
| RTX 5090 32GB | Ollama | Default | 24.5 GB | fits in fast memory | ~273 tok/s | ollama run ornith-1.5:35b |
| M4 Pro 48GB | Ollama | Default | 24.5 GB | fits in fast memory | ~55 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M4 Max (40c GPU) 48GB | Ollama | Default | 24.5 GB | fits in fast memory | ~111 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M5 Max (40c GPU) 48GB | Ollama | Default | 24.5 GB | fits in fast memory | ~125 tok/s | ollama run ornith-1.5:35b |
| M1 Max 64GB | Ollama | Default | 24.5 GB | fits in fast memory | ~81 tok/s | ollama run ornith-1.5:35b |
| M2 Max 64GB | Ollama | Default | 24.5 GB | fits in fast memory | ~81 tok/s | ollama run ornith-1.5:35b |
| M3 Max 64GB | Ollama | Default | 24.5 GB | fits in fast memory | ~81 tok/s | ollama run ornith-1.5:35b |
| M4 Max 64GB | Ollama | Default | 24.5 GB | fits in fast memory | ~111 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M4 Max (40c GPU) 64GB | Ollama | Default | 24.5 GB | fits in fast memory | ~111 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M5 Max (40c GPU) 64GB | Ollama | Default | 24.5 GB | fits in fast memory | ~125 tok/s | ollama run ornith-1.5:35b |
| DGX Spark 64GB (ASUS, Dell, HP, MSI, Acer, Gigabyte) | Ollama | Default | 24.5 GB | fits in fast memory | ~80 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M3 Ultra 96GB | Ollama | Default | 24.5 GB | fits in fast memory | ~166 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M5 Ultra 96GB | Ollama | Default | 24.5 GB | fits in fast memory | ~243 tok/s | ollama run ornith-1.5:35b |
| M3 Max 128GB | Ollama | Default | 24.5 GB | fits in fast memory | ~81 tok/s | ollama run ornith-1.5:35b |
| M4 Max 128GB | Ollama | Default | 24.5 GB | fits in fast memory | ~111 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M4 Max (40c GPU) 128GB | Ollama | Default | 24.5 GB | fits in fast memory | ~111 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M5 Max (40c GPU) 128GB | Ollama | Default | 24.5 GB | fits in fast memory | ~125 tok/s | ollama run ornith-1.5:35b |
| DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte) | Ollama | Default | 24.5 GB | fits in fast memory | ~80 tok/s | ollama run ornith-1.5:35b |
| M2 Ultra 192GB | Ollama | Default | 24.5 GB | fits in fast memory | ~162 tok/s | ollama run ornith-1.5:35b |
| 2x Mac Studio M5 Max (40c GPU) 128GB | llama.cpp | Q8_0 GGUF | 39.3 GB | fits in fast memory | ~64 tok/s | llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q8_0 --rpc node2:50052 -c 8192 # run rpc-server on each other node |
| Mac Studio M3 Ultra 256GB | Ollama | Default | 24.5 GB | fits in fast memory | ~166 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M5 Ultra (64c GPU) 256GB | Ollama | Default | 24.5 GB | fits in fast memory | ~243 tok/s | ollama run ornith-1.5:35b |
| Mac Studio M5 Ultra (80c GPU) 256GB | Ollama | Default | 24.5 GB | fits in fast memory | ~243 tok/s | ollama run ornith-1.5:35b |
| 2x DGX Spark 128GB | llama.cpp | Q8_0 GGUF | 39.3 GB | fits in fast memory | ~41 tok/s | llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q8_0 --rpc node2:50052 -c 8192 # run rpc-server on each other node |
| 4x Mac Studio M5 Max (40c GPU) 128GB | llama.cpp | Q8_0 GGUF | 39.3 GB | fits in fast memory | ~57 tok/s | llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q8_0 --rpc node2:50052,node3:50052,node4:50052 -c 8192 # run rpc-server on each other node |
| 2x Mac Studio M5 Ultra (80c GPU) 256GB | llama.cpp | Q8_0 GGUF | 39.3 GB | fits in fast memory | ~126 tok/s | llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q8_0 --rpc node2:50052 -c 8192 # run rpc-server on each other node |
| Mac Studio M3 Ultra (80c GPU) 512GB | Ollama | Default | 24.5 GB | fits in fast memory | ~166 tok/s | ollama run ornith-1.5:35b |
| 4x DGX Spark 128GB | llama.cpp | Q8_0 GGUF | 39.3 GB | fits in fast memory | ~36 tok/s | llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q8_0 --rpc node2:50052,node3:50052,node4:50052 -c 8192 # run rpc-server on each other node |
| 4x Mac Studio M5 Ultra (80c GPU) 256GB | llama.cpp | Q8_0 GGUF | 39.3 GB | fits in fast memory | ~111 tok/s | llama-server -hf ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q8_0 --rpc node2:50052,node3:50052,node4:50052 -c 8192 # run rpc-server on each other node |
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