arxiv
PublishedJune 12, 2026 at 4:00 AM
Holding the FP8 Quality Ceiling at 8-Bit Weights and Activations: INT8 and GGUF Post-Training Quantization of Ideogram 4.0 for Consumer GPUs
Publisher summary· verbatim
arXiv:2606.12280v1 Announce Type: new Abstract: Post-training quantization lets large text-to-image diffusion transformers run on consumer GPUs, yet the hardware-specific trade-offs are seldom measured directly. We quantize Ideogram 4.0 - a 9.3B flow-matching diffusion transformer (DiT), shipped as
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
Related coverage
More from ARXIV
arxivLoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling5harxivAPPO: Agentic Procedural Policy Optimization5harxivALIGNBEAM : Inference-Time Alignment Transfer via Cross-Vocabulary Logit Mixing5harxivGeoNatureAgent Benchmark: Benchmarking LLM Agents for Environmental Geospatial Analysis Across Frontier and Open-Weight Foundation Models5hThe Bubble Brief
WEEKLYRead AI insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗