Model comparison
Qwen3-Coder-30B-A3B-Instruct vs Olmo-3.1-32B-Instruct
At Q4_K_M, Qwen3-Coder-30B-A3B-Instruct is the smaller download — 18,556,689,568 bytes against 19,482,034,976.
From the file· summed bytes, KV per layer
Side by side
| Qwen3-Coder-30B-A3B-Instruct | Olmo-3.1-32B-Instruct | |
|---|---|---|
| Parameters | 30.5B | 32.2B |
| Architecture | qwen3moe | olmo2 |
| Layers | 48 | 64 |
| Native context | 262,144 | 65,536 |
| Mixture of experts | yes, 128 experts | no |
| Quantizations published | 46 | 49 |
| Smallest quantization | 7.46 GiB | 6.75 GiB |
| Q4_K_M | 17.28 GiB | 18.14 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Qwen3-Coder-30B-A3B-Instruct | Olmo-3.1-32B-Instruct | Ratio |
|---|---|---|---|
| 4,096 | 0.38 GiB | 1.00 GiB | 2.67× |
| 8,192 | 0.75 GiB | 1.34 GiB | 1.79× |
| 16,384 | 1.50 GiB | 1.84 GiB | 1.23× |
| 32,768 | 3.00 GiB | 2.84 GiB | 1.05× |
| 65,536 | 6.00 GiB | 4.84 GiB | 1.24× |
| 131,072 | 12.00 GiB | 8.84 GiB | 1.36× |