[enhancement]: LTX-2 GGUF transformers #9728

Is there an existing issue for this?

• I have searched the existing issues

What should this feature add?

Load community GGUF builds of the LTX-2 (2.5) transformer. Today installation is refused in
configs/main.py with "GGUF-quantized LTX-2 checkpoints are not supported yet"; bf16, Comfy int8_tensorwise and
header/marker-named nvfp4 single files load.

Alternatives

Use the int8 or nvfp4 single file.

Additional Content

• Demand: realrebelai/LTX-2.5_GGUFs, Abiray/LTX-2.5-Distilled-GGUF, vantagewithai/LTX-2.5-GGUF.
• The most downloaded GGUFs (unsloth/LTX-2.3-GGUF, ~371k; QuantStack/LTX-2.3-GGUF) are LTX-2.3, which InvokeAI
  refuses regardless of format (only 2.5 is supported). The refusal message should stay clear for those.
• The pattern from feat(ideogram4): load community GGUF transformers #9708 (Ideogram 4 GGUF) and feat(ernie-image): load GGUF transformers #9725 (ERNIE-Image GGUF) applies: a Main_GGUF_*_Config fingerprinted by keys (GGUF headers often name another architecture), a loader that keeps quantized Linears packed and unpacks the rest with unpack_ggml_at_load, and the denoise node's peak_dequant_transient_bytes, which already counts GGUF layers (fix(model-cache): reserve the GGUF dequantization transient in every node #9709).

The full matrix of what loads today is in #9726 (Model Format Support page).