Model comparison
Wan_2.1_ComfyUI_repackaged vs LTX-2.3
At Q4_K_M, LTX-2.3 is the smaller download — 14,194,441,248 bytes against 68,351,109,664.
From the file· summed bytes, KV per layer
Side by side
| Wan_2.1_ComfyUI_repackaged | LTX-2.3 | |
|---|---|---|
| Parameters | — | 9.2B |
| Architecture | pig | qwen35 |
| Layers | — | — |
| Native context | — | — |
| Mixture of experts | no | no |
| Quantizations published | 48 | 92 |
| Smallest quantization | 4.25 GiB | 7.39 GiB |
| Q4_K_M | 63.66 GiB | 13.22 GiB |
| Licence | apache-2.0 | — |
KV cache by context
the term that decides long-context viability
| Context | Wan_2.1_ComfyUI_repackaged | LTX-2.3 | Ratio |
|---|---|---|---|
| 4,096 | — | — | — |
| 8,192 | — | — | — |
| 16,384 | — | — | — |
| 32,768 | — | — | — |
| 65,536 | — | — | — |
| 131,072 | — | — | — |