Model comparison
Yi-Coder-9B-Chat vs Qwen3-VL-8B-Instruct-abliterated-v1
At Q4_K_M, Qwen3-VL-8B-Instruct-abliterated-v1 is the smaller download — 5,027,785,792 bytes against 5,328,957,984.
From the file· summed bytes, KV per layer
Side by side
| Yi-Coder-9B-Chat | Qwen3-VL-8B-Instruct-abliterated-v1 | |
|---|---|---|
| Parameters | 8.8B | 8.8B |
| Architecture | llama | qwen3vl |
| Layers | 48 | — |
| Native context | 131,072 | — |
| Mixture of experts | no | no |
| Quantizations published | 36 | 48 |
| Smallest quantization | 1.88 GiB | 1.97 GiB |
| Q4_K_M | 4.96 GiB | 4.68 GiB |
| Licence | — | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Yi-Coder-9B-Chat | Qwen3-VL-8B-Instruct-abliterated-v1 | Ratio |
|---|---|---|---|
| 4,096 | 0.38 GiB | — | — |
| 8,192 | 0.75 GiB | — | — |
| 16,384 | 1.50 GiB | — | — |
| 32,768 | 3.00 GiB | — | — |
| 65,536 | 6.00 GiB | — | — |
| 131,072 | 12.00 GiB | — | — |