ModelOpt / FP8¶
ModelOpt and FP8 checkpoint flows are practical quantization families in the codebase. They are useful when a deployment has validated low-precision transformer components or when online quantization is acceptable for the target model.
Role in Sol-Engine¶
The reported full optimization stacks focus on selective NVFP4 for Cosmos3-Super and LTX-2.3, but ModelOpt / FP8 remains part of the broader local quantization search space.
Practical knobs¶
- transformer override path.
- transformer weights path.
- ignored layer patterns.
- layerwise offload compatibility.
- BF16 fallback modules.
Validation¶
Quantized checkpoints should be validated in the same pipeline where they will be composed. A checkpoint that is acceptable alone may become unacceptable after cache or sparse attention is added.