BestLLMfor EN Your hardware. Your LLM. Your call.
APIOpen data Find my LLM
Model fiche

Qwen 3.5 122B-A10B

By Alibaba · China

Updated 2026-07-13

chat general reasoning multilingual moe
Parameters
122B
License
Apache 2.0
Context
255k
VRAM (Q4)
73 GB
Released
April 2025

Overview

Alibaba's mid-flagship Qwen 3.5 with 122B total / 10B active params and 262k native context. Frontier-class quality that fits on a single H100.

When to pick this model

  • Frontier-quality inference on a single H100
  • Long-context document and codebase analysis (262k)
  • Multilingual reasoning workloads
  • Apache 2.0 deployments where Qwen 397B is overkill
  • Cost-sensitive agentic systems needing top-tier quality

VRAM requirements by quantization

VRAM REQUIRED (GB)24324880128Q4_K_M73 GBQ5_K_M88 GBQ8_0131 GBFP16244 GB
QuantizationVRAM required
Q4_K_M (recommended)73 GB
Q5_K_M88 GB
Q8_0131 GB
FP16 (no quantization)244 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

In practice, Qwen 3.5 122B-A10B is server-class even at Q4_K_M (73 GB). Stepping up to Q8_0 nearly doubles the footprint to 131 GB, and unquantized FP16 weights take 244 GB — plan your GPU around the Q4 or Q5 figure unless you specifically need the higher fidelity.

Without a GPU, Qwen 3.5 122B-A10B needs roughly 110 GB of system RAM to run on CPU via llama.cpp or Ollama — workable for background jobs, but far slower than GPU inference. Throughput estimates from our compatibility engine: around 3 tokens/sec on entry-level GPUs, on the order of 12 tokens/sec on a mid-range card, and up to 28 tokens/sec on high-end hardware — assuming the chosen quantization fully fits in VRAM.

What hardware do you need

The table below matches Qwen 3.5 122B-A10B to common GPU memory tiers, using the highest-fidelity quantization that fully fits each card class. Spilling layers to system RAM works but costs most of the speed, so size your card to the quantization you actually want to run.

GPU memoryExample cardsBest fit for Qwen 3.5 122B-A10B
8 GBRTX 5070 Laptop, RTX 5060, RTX 5060 Ti 8GBDoes not fit — needs 73 GB at Q4_K_M
12 GBRTX 5070, RTX 5070 Ti Laptop, RTX 4080 LaptopDoes not fit — needs 73 GB at Q4_K_M
16 GBRTX 5080, RTX 4080 Super, Radeon RX 9070 XTDoes not fit — needs 73 GB at Q4_K_M
24 GBRTX 4090, Radeon RX 7900 XTX, RTX 5090 LaptopDoes not fit — needs 73 GB at Q4_K_M
32 GBRTX 5090Does not fit — needs 73 GB at Q4_K_M

Which GPU should you buy to run Qwen 3.5 122B-A10B?

To run Qwen 3.5 122B-A10B locally at Q4, you need ~73 GB of VRAM. The best value for this is a Apple Mac Studio (64+ GB unified memory).

Check Apple Mac Studio price on Amazon →

As an Amazon Associate, BestLLMfor earns from qualifying purchases, at no extra cost to you. It does not influence our independent rankings.

Strengths

  • Frontier-class quality with only 10B active params
  • 262k native context window
  • Apache 2.0
  • Single-H100 deployment is realistic
  • Strong multilingual coverage

Limitations

  • Roughly 73 GB VRAM in Q4 — still needs multi-GPU on consumer cards
  • Mid-flagship positioning means it's eclipsed by 397B on the hardest tasks

Typical workloads

In our catalog grid, Qwen 3.5 122B-A10B is filed under Workstation MoE, Reasoning — the use cases where its size/quality trade-off makes the most sense. Its tags translate to concrete workloads: multi-step reasoning and math-flavoured tasks; multilingual workloads.

The 255k-token context window is large enough to hold entire codebases' worth of files or long reports in a single prompt, which is what makes local RAG and document analysis practical. The Apache 2.0 license is permissive, so shipping it inside a commercial product raises no special legal questions.

Architecture & training

Architecture: MoE · 122B total / 10B active · Qwen 3.5 flagship · 262k context

Training: Qwen 3.5 accessible flagship — 10B active out of 122B, native 262k ctx.

Verdict

The sweet spot of the Qwen 3.5 lineup: H100-friendly with frontier-grade output.

Quick start

ollama run qwen3.5:122b-a10b

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.

Similar models worth comparing

Frequently asked questions

How much VRAM does Qwen 3.5 122B-A10B need?

At the recommended Q4_K_M quantization, Qwen 3.5 122B-A10B needs about 73 GB of VRAM. Q8_0 takes 131 GB, and unquantized FP16 weights take 244 GB.

Can Qwen 3.5 122B-A10B run without a GPU?

Yes — with roughly 110 GB of system RAM it runs CPU-only through llama.cpp or Ollama. Expect a fraction of GPU speed, which is fine for background or batch jobs but slow for interactive chat.

What context window does Qwen 3.5 122B-A10B support?

Qwen 3.5 122B-A10B supports a 255k-token context window (262,000 tokens).

Can I use Qwen 3.5 122B-A10B commercially?

Yes. Qwen 3.5 122B-A10B is released under Apache 2.0, a permissive open-source license that allows commercial use, modification and redistribution.

How fast is Qwen 3.5 122B-A10B on consumer hardware?

Our compatibility engine estimates on the order of 12 tokens/sec on a mid-range GPU and up to 28 tokens/sec on high-end cards, assuming the quantization fully fits in VRAM.

Which quantization of Qwen 3.5 122B-A10B should I download first?

Start with Q4_K_M (73 GB) — the standard size/quality sweet spot. Step up to Q5_K_M or Q8_0 only if you have VRAM headroom. It does not fit a single 24 GB consumer card — plan for multi-GPU or server hardware.

Tools

Is Qwen 3.5 122B-A10B the right pick for you?

Compute self-hosted ROI → Back to catalog