Gemma 4 mobile-mixture QAT with TQ2_0 2-bit layers (UD-Q2_K_XL for E2B/E4B)
Parent: Mac local LLMs: Quantization formats and methods · Published reference · snapshot 2026-10-05
↓ Facts as markdownall context files
Already covered: gemma-4-qat-gguf-conversion-f16-versus-bf16-scal.md holds the TQ2_0 plus negative-scaler conversion into UD-Q2_K_XL, the E2B (2.19 GB, 61 TQ2_0 tensors, KLD 0.00409) and E4B (3.22 GB, 2 TQ2_0 tensors, KLD 0.00102) rows; no new claim found.
These notes link each claim to its source. A source may be a research report hosted on this site rather than the primary document. A published reference means the content is available; it does not certify independent review or accuracy.Read the editorial policy and follow the sources before relying on a claim.
Facts
- Already covered: gemma-4-qat-gguf-conversion-f16-versus-bf16-scal.md holds the TQ2_0 plus negative-scaler conversion into UD-Q2_K_XL, the E2B (2.19 GB, 61 TQ2_0 tensors, KLD 0.00409) and E4B (3.22 GB, 2 TQ2_0 tensors, KLD 0.00102) rows; no new claim found. [source]
Children
- No children recorded.