Model comparison
LTX-2.3 vs Wan2.2-I2V-A14B
At Q4_K_M, Wan2.2-I2V-A14B is the smaller download — 9,651,728,896 bytes against 14,194,441,248.
From the file· summed bytes, KV per layer
Side by side
| LTX-2.3 | Wan2.2-I2V-A14B | |
|---|---|---|
| Parameters | 9.2B | 14.3B |
| Architecture | qwen35 | wan |
| Layers | — | — |
| Native context | — | — |
| Mixture of experts | no | no |
| Quantizations published | 92 | 30 |
| Smallest quantization | 7.39 GiB | 4.94 GiB |
| Q4_K_M | 13.22 GiB | 8.99 GiB |
| Licence | — | apache-2.0 |
KV cache by context
the term that decides long-context viability
| Context | LTX-2.3 | Wan2.2-I2V-A14B | Ratio |
|---|---|---|---|
| 4,096 | — | — | — |
| 8,192 | — | — | — |
| 16,384 | — | — | — |
| 32,768 | — | — | — |
| 65,536 | — | — | — |
| 131,072 | — | — | — |