Qwen
Qwen
/Qwen3.5-2B

Quantizations

QuantQuantized bySizeDecodePrefillScoreActions
Unsloth
Unsloth
1.5 GBN/AN/AN/A
MLX Community
MLX Community
2.0 GB96.7 tok/s1,095.4 tok/sRuns great

Device Comparison

Results include trials with 4,096 input tokens and 1,024 output tokens only.

Decode / Prefill Speeds

2 devices