Qwen

Qwen3.5 122B A10B

Qwen3.5 122B A10B is an open-weights 122B-parameter model by Qwen. It is listed as in-progress on the local.ai leaderboard.

MoEopen weights
Parameters
122B
MoE · 10B active parameters
Benchmarks (local.ai snapshot)
τ²-bench
GAIA
GDPval
HuggingFace metadata · updated 2026-04-24
repo
Qwen/Qwen3.5-122B-A10B
license
apache-2.0
downloads
2,220,760
likes
606

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 70.8GQ6_K 109.8GQ8_0 134.2GFP16 244G
Quantization≈ File sizeNote
IQ2_XXS32.9 GBExtreme compression, largest quality loss
IQ3_XXS43.9 GBVery small
Q3_K_M56.1 GBSmall
NVFP461 GBNVIDIA 4-bit
Q4_K_S65.9 GBCommon
Q4_K_M70.8 GBMost common sweet spot
Q5_K_M92.7 GBBalanced
Q6_K109.8 GBNear-lossless
Q8_0134.2 GBNear-lossless
FP16244 GBOriginal weights

Actual files on HuggingFace

FileSize
model.safetensors-00025-of-00039.safetensors6.00 GB
model.safetensors-00026-of-00039.safetensors6.00 GB
model.safetensors-00027-of-00039.safetensors6.00 GB
model.safetensors-00028-of-00039.safetensors6.00 GB
model.safetensors-00029-of-00039.safetensors6.00 GB
model.safetensors-00030-of-00039.safetensors6.00 GB
GGUF: unsloth/Qwen3.5-122B-A10B-GGUF · 307,349 downloads
FileSize
BF16/Qwen3.5-122B-A10B-BF16-00001-of-00005.gguf46.51 GBdownload
BF16/Qwen3.5-122B-A10B-BF16-00002-of-00005.gguf46.50 GBdownload
BF16/Qwen3.5-122B-A10B-BF16-00003-of-00005.gguf46.50 GBdownload
BF16/Qwen3.5-122B-A10B-BF16-00004-of-00005.gguf46.50 GBdownload
BF16/Qwen3.5-122B-A10B-BF16-00005-of-00005.gguf41.52 GBdownload
MXFP4_MOE/Qwen3.5-122B-A10B-MXFP4_MOE-00001-of-00003.gguf0.01 GBdownload
MXFP4_MOE/Qwen3.5-122B-A10B-MXFP4_MOE-00002-of-00003.gguf46.23 GBdownload
MXFP4_MOE/Qwen3.5-122B-A10B-MXFP4_MOE-00003-of-00003.gguf23.30 GBdownload
Q3_K_M/Qwen3.5-122B-A10B-Q3_K_M-00001-of-00003.gguf0.01 GBdownload
Q3_K_M/Qwen3.5-122B-A10B-Q3_K_M-00002-of-00003.gguf46.18 GBdownload
Q3_K_M/Qwen3.5-122B-A10B-Q3_K_M-00003-of-00003.gguf6.35 GBdownload
Q3_K_S/Qwen3.5-122B-A10B-Q3_K_S-00001-of-00003.gguf0.01 GBdownload
GGUF: unsloth/Qwen3.5-122B-A10B-MTP-GGUF · 167,867 downloads
FileSize
BF16/Qwen3.5-122B-A10B-BF16-00001-of-00005.gguf46.51 GBdownload
BF16/Qwen3.5-122B-A10B-BF16-00002-of-00005.gguf46.50 GBdownload
BF16/Qwen3.5-122B-A10B-BF16-00003-of-00005.gguf46.50 GBdownload
BF16/Qwen3.5-122B-A10B-BF16-00004-of-00005.gguf46.50 GBdownload
BF16/Qwen3.5-122B-A10B-BF16-00005-of-00005.gguf46.23 GBdownload
MXFP4_MOE/Qwen3.5-122B-A10B-MXFP4_MOE-00001-of-00003.gguf0.01 GBdownload
MXFP4_MOE/Qwen3.5-122B-A10B-MXFP4_MOE-00002-of-00003.gguf46.26 GBdownload
MXFP4_MOE/Qwen3.5-122B-A10B-MXFP4_MOE-00003-of-00003.gguf24.73 GBdownload
Q8_0/Qwen3.5-122B-A10B-Q8_0-00001-of-00004.gguf0.01 GBdownload
Q8_0/Qwen3.5-122B-A10B-Q8_0-00002-of-00004.gguf46.36 GBdownload
Q8_0/Qwen3.5-122B-A10B-Q8_0-00003-of-00004.gguf46.39 GBdownload
Q8_0/Qwen3.5-122B-A10B-Q8_0-00004-of-00004.gguf30.69 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.