Qwen

Qwen3.6 27B

Qwen3.6 27B is an open-weights 27B-parameter model by Qwen. It holds an Intelligence score of 76.5 on the local.ai leaderboard (2026-08 snapshot).

open weights
76.5
local.ai Intelligence
Parameters
27B
Benchmarks (local.ai snapshot)
τ²-bench
100%
GAIA
74%
GDPval
74%
HuggingFace metadata · updated 2026-04-24
repo
Qwen/Qwen3.6-27B
license
apache-2.0
downloads
7,050,183
likes
2,263

Metadata refreshed from the HuggingFace API (snapshot 2026-08-15T15:53:42.405Z). Context length and long-form descriptions are the next pipeline stage.

Does it fit your hardware?

Detect or select your GPU first. Green = weights + context headroom, yellow = tight, grey = does not fit.

Q4_K_M 15.7GQ6_K 24.3GQ8_0 29.7GFP16 54G
Quantization≈ File sizeNote
IQ2_XXS7.3 GBExtreme compression, largest quality loss
IQ3_XXS9.7 GBVery small
Q3_K_M12.4 GBSmall
NVFP413.5 GBNVIDIA 4-bit
Q4_K_S14.6 GBCommon
Q4_K_M15.7 GBMost common sweet spot
Q5_K_M20.5 GBBalanced
Q6_K24.3 GBNear-lossless
Q8_029.7 GBNear-lossless
FP1654 GBOriginal weights

Actual files on HuggingFace

FileSize
model-00013-of-00015.safetensors3.72 GB
model-00007-of-00015.safetensors3.72 GB
model-00001-of-00015.safetensors3.70 GB
model-00014-of-00015.safetensors3.67 GB
model-00002-of-00015.safetensors3.65 GB
model-00009-of-00015.safetensors3.65 GB
GGUF: DavidAU/Qwen3.6-27B-Fable-Fusion-711-Uncensored-Heretic-NM-DAU-NEO-MAX-MTP-GGUF · 2,983,500 downloads
FileSize
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-AMD-MTP-IQ4_XS.gguf15.65 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-AMD-MTP-Q6_K.gguf22.38 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-IQ2_M.gguf10.87 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-IQ3_M.gguf13.11 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-IQ4_NL.gguf16.11 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-IQ4_XS.gguf15.44 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-LOW-MTP-IQ4_XS.gguf14.10 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-LOW-MTP-Q6_K.gguf21.27 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-MTP-IQ2_M.gguf11.29 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-MTP-IQ3_M.gguf13.53 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-MTP-IQ4_NL.gguf16.53 GBdownload
Qwen3.6-27B-Fable-Fus-711-UnHeretic-NM-DAU-NEO-MAX-NEO-MTP-IQ4_XS.gguf15.86 GBdownload
GGUF: unsloth/Qwen3.6-27B-MTP-GGUF · 1,217,421 downloads
FileSize
BF16/Qwen3.6-27B-BF16-00001-of-00002.gguf46.45 GBdownload
BF16/Qwen3.6-27B-BF16-00002-of-00002.gguf4.45 GBdownload
Qwen3.6-27B-IQ4_NL.gguf15.22 GBdownload
Qwen3.6-27B-IQ4_XS.gguf14.63 GBdownload
Qwen3.6-27B-Q3_K_M.gguf12.87 GBdownload
Qwen3.6-27B-Q3_K_S.gguf11.71 GBdownload
Qwen3.6-27B-Q4_0.gguf14.95 GBdownload
Qwen3.6-27B-Q4_1.gguf16.34 GBdownload
Qwen3.6-27B-Q4_K_M.gguf15.93 GBdownload
Qwen3.6-27B-Q4_K_S.gguf15.01 GBdownload
Qwen3.6-27B-Q5_K_M.gguf18.47 GBdownload
Qwen3.6-27B-Q5_K_S.gguf17.95 GBdownload

What you save self-hosting this model

Uses the model's own API price when it exists; otherwise the closest commercial equivalent.

Waiting for hardware detection…

RunLocal presents third-party leaderboard data (local.ai / Exo Labs, 2026-08) with its own fit and cost estimates. Verify pricing, licenses and file sizes at the source before deploying a model.