# Qwen3.5-27B dense Unsloth KLD table for dense versus MoE elasticity > The Unsloth Qwen3.5 GGUF benchmarks page links the Qwen3.5-27B GGUF repo but its benchmark charts and tables are for 35B-A3B and 122B-A10B; the page states new benchmarks for 122B-A10B and 35B-A3B. Parent: [Mac local LLMs: Quantization evaluation](https://llms-explorer.com/tree/mac-local-llms-quantization-evaluation/) · 2 facets · 14 facts · page: https://llms-explorer.com/tree/qwen3-5-27b-dense-unsloth-kld-table-for-dense-ve/ ## Facts - The Unsloth Qwen3.5 GGUF benchmarks page links the Qwen3.5-27B GGUF repo but its benchmark charts and tables are for 35B-A3B and 122B-A10B; the page states new benchmarks for 122B-A10B and 35B-A3B. — [source](https://unsloth.ai/docs/models/qwen3.5/gguf-benchmarks) - localbench's Qwen 3.6 27B post puts Q8_0 KL at 0.075, against 0.069 for Qwen 3.6 35B-A3B and 0.120 for Qwen 3.5 27B, all on its 250K-token mixed set. — [source](https://localbench.substack.com/p/qwen-3-6-27b-gguf-quality-benchmark) - On that set the dense Qwen 3.5 27B Q8_0 floor is about 1.7 times the Qwen 3.6 27B floor, so a dense model can carry a higher irreducible KLD at near-lossless size. — source: `asserted` - 5 Mar 2026: Unsloth reissued Qwen3.5 35B, 27B, 122B and 397B with a new quantization algorithm and new imatrix data and asked users to redownload; earlier 27B numbers describe superseded files. — [source](https://unsloth.ai/docs/models/qwen3.5/gguf-benchmarks) - Unsloth's tensor findings (do not quantize ssm_out; ffn_down_exps is sensitive) are written for the MoE hybrid; a dense 27B has no expert tensors, so those rules do not transfer directly. — [source](https://unsloth.ai/docs/models/qwen3.5/gguf-benchmarks) - localbench KL and Unsloth KLD use different datasets and estimators, so the 0.120 Q8_0 value cannot be placed on the Unsloth GB axis. — [source](https://localbench.substack.com/p/gguf-benchmark-methodology) - Unsloth's own guidance is to use 27B when accuracy matters and 35B-A3B when speed matters, which is a quality claim without a published 27B KLD row. — [source](https://unsloth.ai/docs/models/qwen3.5) - Unsloth's Experiments repo may hold 27B KLD files, but the page text does not list 27B there. — source: `asserted` - The dense-versus-MoE elasticity stays unmeasured; the nearest data is localbench per-quant KL for Qwen 3.6 27B (87 quants), whose tables are images. — [source](https://localbench.substack.com/p/qwen-3-6-27b-gguf-quality-benchmark) - The Unsloth Qwen3.5 GGUF benchmark page published no Qwen3.5-27B KLD table as of the cached copy (2026-10-04). — [source](https://unsloth.ai/docs/models/qwen3.5/gguf-benchmarks) - localbench Q8_0 KL: Qwen 3.5 27B 0.120, Qwen 3.6 27B 0.075, Qwen 3.6 35B-A3B 0.069. — [source](https://localbench.substack.com/p/qwen-3-6-27b-gguf-quality-benchmark) - Unsloth's 5 Mar 2026 update replaced the Qwen3.5 27B GGUFs with a new algorithm and imatrix. — [source](https://unsloth.ai/docs/models/qwen3.5/gguf-benchmarks) - The dense-versus-MoE elasticity question has no published table to answer it. — source: `asserted` ## Corrections and disagreements - CONTRADICTS: log-log-size-to-kld-elasticity-of-k-quants-acros.md Open questions line "Qwen3.5-27B has an Unsloth table not read here": the cached 2026-10-04 page has no 27B KLD rows to read. — [source](https://unsloth.ai/docs/models/qwen3.5/gguf-benchmarks)