LocalOps
Back to Calculator

Qwen3.6 35B-A3B

First open-weight Qwen3.6 model. 35B total / 3B active MoE, focused on agentic coding and repo-level reasoning. Native 262K context, extensible to 1M tokens. Apache 2.0.

Model Specifications

ArchitectureTEXT
Parameters35B
Familyqwen3.6
VRAM (Q4)17.5GB
Mixture of ExpertsActive inference parameters: 3B.
#coding#agents#thinkingSource

Estimated Quantization Sizes

FormatPrecisionEst. VRAMRecommendation
FP16 / BF1616-bit70.0 GBUncompressed Base
Q8_0High8-bit35.0 GBNear Lossless
Q6_K6-bit26.3 GBExcellent Balance
Q4_K_MPopular4-bit17.5 GBStandard Use

Share this Model

Send this model's specs directly to your community.

Post