Model comparison
Wan2.2-TI2V-5B vs Wan2.2-TI2V-5B-Turbo
At Q4_K_M, Wan2.2-TI2V-5B is the smaller download — 3,433,116,000 bytes against 3,437,927,136.
From the file· summed bytes, KV per layer
Side by side
| Wan2.2-TI2V-5B | Wan2.2-TI2V-5B-Turbo | |
|---|---|---|
| Parameters | 5.0B | 5.0B |
| Architecture | wan | wan |
| Layers | — | — |
| Native context | — | — |
| Mixture of experts | no | no |
| Quantizations published | 39 | 22 |
| Smallest quantization | 1.73 GiB | 1.73 GiB |
| Q4_K_M | 3.20 GiB | 3.20 GiB |
| Licence | apache-2.0 | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | Wan2.2-TI2V-5B | Wan2.2-TI2V-5B-Turbo | Ratio |
|---|---|---|---|
| 4,096 | — | — | — |
| 8,192 | — | — | — |
| 16,384 | — | — | — |
| 32,768 | — | — | — |
| 65,536 | — | — | — |
| 131,072 | — | — | — |