Unverified50% confidenceFactExact time
LTX 2.5 splits model components into separate files and directly provides INT8 and NVFP4 quantization formats.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/10/2026
First Seen
Valid until: 12/9/2026
Sources
Related Entities
Related Claims
UnverifiedUnsloth Studio now supports exporting multiple quantization formats after training, including NVFP4, FP8, and imatrix GGUF.62% similarUnverifiedDocker Model Runner使用GGUF格式的量化模型文件，这些文件被打包为符合OCI规范的artifact60% similarUnverifiedLTX-2.3 基于扩散变换器（DiT）架构，专为高效本地推理设计60% similarVerifiedModel weights are typically distributed in GGUF format, a model file format defined by the llama.cpp project and optimized for CPU/GPU hybrid inference.60% similarUnverifiedNVFP4是NVIDIA私有的FP4实现，MXFP8是微软MX浮点标准的FP8变体，二者为Blackwell架构专属格式60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/890281API
curl https://kongchang.com/api/v1/knowledge/claims/890281MCP
get_claim(id=890281)