head to head · same machine, same method
Qwen2.5-0.5B-Instruct-4bit vs SmolLM2-135M-Instruct
Two tiny text-generation models compared on the facts this hub states: what it is, how big it is, and how long one fixed input took on one machine. No benchmark scores from publishers — those live, with their conditions, on each model page.
| mlx-community/Qwen2.5-0.5B-Instruct-4bit | HuggingFaceTB/SmolLM2-135M-Instruct | |
|---|---|---|
| task | text-generation | text-generation |
| format | mlx | safetensors |
| parameters | 494M | 135M |
| file size | 278 MB | 269 MB |
| licence | apache-2.0 | apache-2.0 |
| publisher | mlx-community | HuggingFaceTB |
| one fixed input (ms) | 241.4 | 110.2 |
| downloads | 20 | 21 |
The millisecond row is one fixed input per task type, measured on this hub's machine with one warm-up run then one timed run. How these numbers were taken. Saved filenames: Qwen2.5-0.5B-Instruct-4bit.safetensors · SmolLM2-135M-Instruct.safetensors.