Model comparison
Nemotron-Research-Reasoning-Qwen-1.5B vs Qwen3-4B
At Q4_K_M, Nemotron-Research-Reasoning-Qwen-1.5B is the smaller download — 1,117,321,248 bytes against 2,497,280,256. At long context the gap widens: Nemotron-Research-Reasoning-Qwen-1.5B's KV cache at 32K is 5.1× smaller, which usually matters more than the difference in weights.
From the file· summed bytes, KV per layer
Side by side
| Nemotron-Research-Reasoning-Qwen-1.5B | Qwen3-4B | |
|---|---|---|
| Parameters | 1.8B | 4.0B |
| Architecture | qwen2 | qwen3 |
| Layers | 28 | 36 |
| Native context | 131,072 | 40,960 |
| Mixture of experts | no | no |
| Quantizations published | 23 | 29 |
| Smallest quantization | 0.65 GiB | 1.01 GiB |
| Q4_K_M | 1.04 GiB | 2.33 GiB |
| Licence | cc-by-nc-4.0 | — |
KV cache by context
the term that decides long-context viability
| Context | Nemotron-Research-Reasoning-Qwen-1.5B | Qwen3-4B | Ratio |
|---|---|---|---|
| 4,096 | 0.11 GiB | 0.56 GiB | 5.14× |
| 8,192 | 0.22 GiB | 1.13 GiB | 5.14× |
| 16,384 | 0.44 GiB | 2.25 GiB | 5.14× |
| 32,768 | 0.88 GiB | 4.50 GiB | 5.14× |
| 65,536 | 1.75 GiB | 9.00 GiB | 5.14× |
| 131,072 | 3.50 GiB | 18.00 GiB | 5.14× |