← CanIRunLocal · All hardware · All models

Can I run Qwen3.5 0.8B locally?

Alibaba · 0.87B parameters · chat, reasoning, coding, vision.

Ollama weight files: Q8_0 1 GB (qwen3.5:0.8b-q8_0).

Yes on 62 machines in this list. The smallest is M1 8GB: Q8_0, about 2 GB, roughly 54 tokens per second, via Ollama. ollama run qwen3.5:0.8b-q8_0

DGX Spark 128GB: Q8_0, ~2 GB, fits in fast memory, ~205 tok/s — ollama run qwen3.5:0.8b-q8_0
Mac Studio M5 Max 128GB: Q8_0, ~2 GB, fits in fast memory, ~491 tok/s — ollama run qwen3.5:0.8b-q8_0
Mac Studio M5 Ultra 256GB: Q8_0, ~2 GB, fits in fast memory, ~960 tok/s — ollama run qwen3.5:0.8b-q8_0

Setup on DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte)

Ollama: ollama run qwen3.5:0.8b-q8_0
LM Studio: lms get unsloth/Qwen3.5-0.8B-GGUF@Q8_0
llama.cpp: llama-server -hf unsloth/Qwen3.5-0.8B-GGUF:Q8_0 -c 8192
vLLM: vllm serve Qwen/Qwen3.5-0.8B --max-model-len 8192 --gpu-memory-utilization 0.9

Open Qwen3.5 0.8B in the checker

Hardware that fits

Hardware	Runtime	Quant	Memory	Fit	Speed	Command
M1 8GB	Ollama	Q8_0	2 GB	fits in fast memory	~54 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M2 8GB	Ollama	Q8_0	2 GB	fits in fast memory	~80 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M3 8GB	Ollama	Q8_0	2 GB	fits in fast memory	~80 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 3060 Ti 8GB	Ollama	Q8_0	2 GB	fits in fast memory	~269 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 3070 8GB	Ollama	Q8_0	2 GB	fits in fast memory	~269 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 4060 8GB	Ollama	Q8_0	2 GB	fits in fast memory	~163 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RX 7600 8GB	Ollama	Q8_0	2 GB	fits in fast memory	~173 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 3080 10GB	Ollama	Q8_0	2 GB	fits in fast memory	~456 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M1 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~54 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M2 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~80 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M3 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~80 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M4 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~96 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 3060 12GB	Ollama	Q8_0	2 GB	fits in fast memory	~216 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 4070 12GB	Ollama	Q8_0	2 GB	fits in fast memory	~302 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 5070 12GB	Ollama	Q8_0	2 GB	fits in fast memory	~403 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 4060 Ti 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~173 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 4070 Ti Super 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~403 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 4080 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~430 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 5080 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~576 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RX 7800 XT 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~374 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RX 9070 XT 16GB	Ollama	Q8_0	2 GB	fits in fast memory	~384 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M2 24GB	Ollama	Q8_0	2 GB	fits in fast memory	~80 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M3 24GB	Ollama	Q8_0	2 GB	fits in fast memory	~80 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M4 24GB	Ollama	Q8_0	2 GB	fits in fast memory	~96 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 3090 24GB	Ollama	Q8_0	2 GB	fits in fast memory	~562 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 4090 24GB	Ollama	Q8_0	2 GB	fits in fast memory	~605 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RX 7900 XTX 24GB	Ollama	Q8_0	2 GB	fits in fast memory	~576 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M1 Pro 32GB	Ollama	Q8_0	2 GB	fits in fast memory	~160 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M2 Pro 32GB	Ollama	Q8_0	2 GB	fits in fast memory	~160 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M4 32GB	Ollama	Q8_0	2 GB	fits in fast memory	~96 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M3 Pro 36GB	Ollama	Q8_0	2 GB	fits in fast memory	~120 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M4 Max (32c GPU) 36GB	Ollama	Q8_0	2 GB	fits in fast memory	~328 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M5 Max (32c GPU) 36GB	Ollama	Q8_0	2 GB	fits in fast memory	~368 tok/s	`ollama run qwen3.5:0.8b-q8_0`
RTX 5090 32GB	Ollama	Q8_0	2 GB	fits in fast memory	~1075 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M4 Pro 48GB	Ollama	Q8_0	2 GB	fits in fast memory	~218 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M4 Max (40c GPU) 48GB	Ollama	Q8_0	2 GB	fits in fast memory	~437 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M5 Max (40c GPU) 48GB	Ollama	Q8_0	2 GB	fits in fast memory	~491 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M1 Max 64GB	Ollama	Q8_0	2 GB	fits in fast memory	~320 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M2 Max 64GB	Ollama	Q8_0	2 GB	fits in fast memory	~320 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M3 Max 64GB	Ollama	Q8_0	2 GB	fits in fast memory	~320 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M4 Max 64GB	Ollama	Q8_0	2 GB	fits in fast memory	~437 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M4 Max (40c GPU) 64GB	Ollama	Q8_0	2 GB	fits in fast memory	~437 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M5 Max (40c GPU) 64GB	Ollama	Q8_0	2 GB	fits in fast memory	~491 tok/s	`ollama run qwen3.5:0.8b-q8_0`
DGX Spark 64GB (ASUS, Dell, HP, MSI, Acer, Gigabyte)	Ollama	Q8_0	2 GB	fits in fast memory	~205 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M3 Ultra 96GB	Ollama	Q8_0	2 GB	fits in fast memory	~655 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M5 Ultra 96GB	Ollama	Q8_0	2 GB	fits in fast memory	~960 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M3 Max 128GB	Ollama	Q8_0	2 GB	fits in fast memory	~320 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M4 Max 128GB	Ollama	Q8_0	2 GB	fits in fast memory	~437 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M4 Max (40c GPU) 128GB	Ollama	Q8_0	2 GB	fits in fast memory	~437 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M5 Max (40c GPU) 128GB	Ollama	Q8_0	2 GB	fits in fast memory	~491 tok/s	`ollama run qwen3.5:0.8b-q8_0`
DGX Spark 128GB (NVIDIA, ASUS, Dell, Lenovo, HP, MSI, Acer, Gigabyte)	Ollama	Q8_0	2 GB	fits in fast memory	~205 tok/s	`ollama run qwen3.5:0.8b-q8_0`
M2 Ultra 192GB	Ollama	Q8_0	2 GB	fits in fast memory	~640 tok/s	`ollama run qwen3.5:0.8b-q8_0`
2x Mac Studio M5 Max (40c GPU) 128GB	llama.cpp	Q8_0 GGUF	1.8 GB	fits in fast memory	~515 tok/s	`llama-server -hf unsloth/Qwen3.5-0.8B-GGUF:Q8_0 --rpc node2:50052 -c 8192 # run rpc-server on each other node`
Mac Studio M3 Ultra 256GB	Ollama	Q8_0	2 GB	fits in fast memory	~655 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M5 Ultra (64c GPU) 256GB	Ollama	Q8_0	2 GB	fits in fast memory	~960 tok/s	`ollama run qwen3.5:0.8b-q8_0`
Mac Studio M5 Ultra (80c GPU) 256GB	Ollama	Q8_0	2 GB	fits in fast memory	~960 tok/s	`ollama run qwen3.5:0.8b-q8_0`
2x DGX Spark 128GB	llama.cpp	Q8_0 GGUF	1.8 GB	fits in fast memory	~215 tok/s	`llama-server -hf unsloth/Qwen3.5-0.8B-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	1.8 GB	fits in fast memory	~455 tok/s	`llama-server -hf unsloth/Qwen3.5-0.8B-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	1.8 GB	fits in fast memory	~1007 tok/s	`llama-server -hf unsloth/Qwen3.5-0.8B-GGUF:Q8_0 --rpc node2:50052 -c 8192 # run rpc-server on each other node`
Mac Studio M3 Ultra (80c GPU) 512GB	Ollama	Q8_0	2 GB	fits in fast memory	~655 tok/s	`ollama run qwen3.5:0.8b-q8_0`
4x DGX Spark 128GB	llama.cpp	Q8_0 GGUF	1.8 GB	fits in fast memory	~190 tok/s	`llama-server -hf unsloth/Qwen3.5-0.8B-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	1.8 GB	fits in fast memory	~889 tok/s	`llama-server -hf unsloth/Qwen3.5-0.8B-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 2026-10-18.

A project by OnDevice Labs · llms.txt