Qwen

Qwen3.5 9B Base

Qwen3.5 9B Base is an open-weights 9B-parameter model by Qwen. It holds an Intelligence score of 59.3 on the local.ai leaderboard (2026-08 snapshot).

open weights
59.3
local.ai Intelligence
Parameters
9B
Benchmarks (local.ai snapshot)
τ²-bench
99%
GAIA
54%
GDPval
55%
HuggingFace metadata · updated 2026-04-23
repo
Qwen/Qwen3.5-9B-Base
license
apache-2.0
downloads
341,161
likes
98

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 5.2GQ6_K 8.1GQ8_0 9.9GFP16 18G
Quantization≈ File sizeNote
IQ2_XXS2.4 GBExtreme compression, largest quality loss
IQ3_XXS3.2 GBVery small
Q3_K_M4.1 GBSmall
NVFP44.5 GBNVIDIA 4-bit
Q4_K_S4.9 GBCommon
Q4_K_M5.2 GBMost common sweet spot
Q5_K_M6.8 GBBalanced
Q6_K8.1 GBNear-lossless
Q8_09.9 GBNear-lossless
FP1618 GBOriginal weights

Actual files on HuggingFace

FileSize
model.safetensors-00003-of-00004.safetensors5.00 GB
model.safetensors-00002-of-00004.safetensors4.97 GB
model.safetensors-00001-of-00004.safetensors4.91 GB
model.safetensors-00004-of-00004.safetensors3.10 GB
GGUF: mradermacher/Qwen3.5-9B-Base-i1-GGUF · 737 downloads
FileSize
Qwen3.5-9B-Base.i1-IQ1_M.gguf2.68 GBdownload
Qwen3.5-9B-Base.i1-IQ1_S.gguf2.55 GBdownload
Qwen3.5-9B-Base.i1-IQ2_M.gguf3.36 GBdownload
Qwen3.5-9B-Base.i1-IQ2_S.gguf3.19 GBdownload
Qwen3.5-9B-Base.i1-IQ2_XS.gguf3.06 GBdownload
Qwen3.5-9B-Base.i1-IQ2_XXS.gguf2.89 GBdownload
Qwen3.5-9B-Base.i1-IQ3_M.gguf4.11 GBdownload
Qwen3.5-9B-Base.i1-IQ3_S.gguf4.07 GBdownload
Qwen3.5-9B-Base.i1-IQ3_XS.gguf3.95 GBdownload
Qwen3.5-9B-Base.i1-IQ3_XXS.gguf3.67 GBdownload
Qwen3.5-9B-Base.i1-IQ4_NL.gguf5.05 GBdownload
Qwen3.5-9B-Base.i1-IQ4_XS.gguf4.84 GBdownload
GGUF: mradermacher/Qwen3.5-9B-Base-GGUF · 398 downloads
FileSize
Qwen3.5-9B-Base.IQ4_XS.gguf4.87 GBdownload
Qwen3.5-9B-Base.Q2_K.gguf3.56 GBdownload
Qwen3.5-9B-Base.Q3_K_L.gguf4.59 GBdownload
Qwen3.5-9B-Base.Q3_K_M.gguf4.31 GBdownload
Qwen3.5-9B-Base.Q3_K_S.gguf3.97 GBdownload
Qwen3.5-9B-Base.Q4_K_M.gguf5.24 GBdownload
Qwen3.5-9B-Base.Q4_K_S.gguf4.98 GBdownload
Qwen3.5-9B-Base.Q5_K_M.gguf6.02 GBdownload
Qwen3.5-9B-Base.Q5_K_S.gguf5.87 GBdownload
Qwen3.5-9B-Base.Q6_K.gguf6.85 GBdownload
Qwen3.5-9B-Base.Q8_0.gguf8.87 GBdownload
Qwen3.5-9B-Base.f16.gguf16.69 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.