Skip to content

Qwen 3.5 122B

by Alibaba · qwen-3.5 family

122B

parameters

text-generation code-generation reasoning multilingual vision tool-use math creative-writing summarization

Qwen 3.5 122B is Alibaba's largest locally-runnable Qwen 3.5 model, delivering frontier-class performance across reasoning, coding, vision, and agentic tasks. With 256K context and 201 languages, it's among the most capable open models available. At 85 GB VRAM (Q4), it requires multi-GPU setups or high-memory Apple Silicon Macs (96 GB+). Best suited for users with workstation-class hardware who want maximum local AI capability.

Quick Start with Ollama

ollama run 122b-q4_K_M
Resources Ollama Hugging Face Official Page
Creator Alibaba
Parameters 122B
Architecture transformer-decoder
Context 256K tokens
Released Mar 15, 2026
License Apache 2.0
Ollama qwen3.5:122b

Quantization Options

Format File Size VRAM Required Quality Ollama Tag
Q4_K_M rec 81 GB 85 GB 122b-q4_K_M
Q8_0 130 GB 135 GB 122b-q8_0

Compatible Hardware

Q4_K_M requires 85 GB VRAM

Compatible Hardware

HardwareVRAMTypeFitEst. Speed
Mac Studio M4 Ultra 512GB512 GBmacRuns~10 tok/s
Mac Pro M2 Ultra 192GB192 GBmacRuns~9 tok/s
Mac Studio M4 Ultra 192GB192 GBmacRuns~10 tok/s
Mac Studio M4 Max 128GB128 GBmacRuns~6 tok/s
MacBook Pro M4 Max 128GB128 GBmacRuns~6 tok/s
MacBook Pro M5 Max 128GB128 GBmacRuns~6 tok/s
NVIDIA RTX PRO 6000 Blackwell96 GBgpuRuns (tight)~23 tok/s
MacBook Pro M3 Max 96GB96 GBmacRuns (tight)~5 tok/s
Mac mini M4 Pro 64GB64 GBmacCPU Offload~1 tok/s
Mac Studio M4 Max 64GB64 GBmacCPU Offload~2 tok/s
MacBook Pro M4 Max 64GB64 GBmacCPU Offload~2 tok/s
MacBook Pro M5 Max 64GB64 GBmacCPU Offload~2 tok/s
95 hardware device(s) cannot run this model at Q4_K_M.