Skip to content
Local AI Wizard
Your machine
Theme

Qwen3 family: which size to run

5 sizes, 1.7B–30.5B parameters · Alibaba

Checked

Sizes

Qwen3 sizes
ModelArchitecture
Qwen3 1.7BAlibaba · Apache 2.0 · Apr 29, 20251.7BDense32,768No LiveBench result for this model in release 2026-06-251041.11
Qwen3 4B Instruct 2507Alibaba · Apache 2.0 · Aug 6, 20254BDense262,144No LiveBench result for this model in release 2026-06-251152.52.3
Qwen3 8BAlibaba · Apache 2.0 · Apr 29, 20258.2BDense32,768No LiveBench result for this model in release 2026-06-2515954.6
Qwen3 14BAlibaba · Apache 2.0 · Apr 29, 202514.8BDense32,768No LiveBench result for this model in release 2026-06-25201498.3
Qwen3 30B A3B Instruct 2507Alibaba · Apache 2.0 · Jul 30, 202530.5BMoE · 3.3B active262,144No LiveBench result for this model in release 2026-06-25312518.617.2

Each size is its own model, with its own scores and memory needs. Q4 estimates are planning figures, not measurements; file sizes are published. The Mac or PC column is for a 32 GB or larger machine (8 GB reserve); machines under 32 GB reserve 25% of their memory (2 to 8 GB), so they need less. The publisher lists 0.6B, 1.7B, 4B, 8B, 14B, 32B (dense); 30B-A3B, 235B-A22B (MoE).

The 4B (August 2025) and 30B A3B (July 2025) here are the “2507” updates (Instruct, non-thinking) of those two sizes; 1.7B, 8B and 14B are the original April 2025 release. Each is its own checkpoint with its own scores.

How to read this table

Q4 means compressed to about 4 bits per weight. GGUF (llama.cpp, Ollama, LM Studio) and MLX (Apple silicon) are file formats; the sizes shown are the published 4-bit files. A dash means no value is published or collected for that size; the reason is in the cell's accessible text.

Which size fits my machine

The largest size whose Q4 planning estimate fits what each machine offers. A planning estimate, not a measurement and not a speed.

Largest Qwen3 size per machine
MachineLargest Qwen3 size with enough memory
Mac mini M2 · 8 GB unified8Qwen3 4B Instruct 25075
Mac mini M4 · 16 GB unified16Qwen3 14B16
Mac mini M4 · 24 GB unified24Qwen3 14B18
Mac mini M4 · 32 GB unified32Qwen3 30B A3B Instruct 250731
Mac mini M5 Pro · 64 GB unified56Qwen3 30B A3B Instruct 250731
Mac Studio M5 Max · 128 GB unified104Qwen3 30B A3B Instruct 250731
Mac Studio · M5 Ultra · 36-core CPU / 80-core GPU · 512 GB unified392Qwen3 30B A3B Instruct 250731
RTX 4060 · 8 GB VRAM8Qwen3 4B Instruct 25075
RTX 3060 · 12 GB VRAM12Qwen3 8B9
RTX 5080 · 16 GB VRAM16Qwen3 14B14
RTX 4090 · 24 GB VRAM24Qwen3 14B14
RTX 5090 · 32 GB VRAM32Qwen3 30B A3B Instruct 250725
DGX H100 · 8 × 80 GB640Qwen3 30B A3B Instruct 250739

A dash means no size of this family fits that machine. The requirement is the size's Q4 estimate with that machine's own reserve: a graphics card reserves 2 GB, as in the sizes table; a Mac or PC reserves 25% of its memory (2 to 8 GB), so below 32 GB its figure is lower than the sizes table's. A server profile reserves 2 GB per graphics card, so its figure is higher.

Smallest machine for each size

The profile with the least memory that still has enough for the size's Q4 estimate, from the hardware catalog.

Smallest machine per Qwen3 size
SizeSmallest Mac or PC-memory profileSmallest graphics-card profile
Qwen3 1.7BMac mini 2020 · M1 · 8 GBGTX 1650 · 4 GB
Qwen3 4B Instruct 2507Mac mini 2020 · M1 · 8 GBRTX 3050 · 6 GB
Qwen3 8BMac mini 2020 · M1 · 16 GBRTX 3060 · 12 GB
Qwen3 14BMac mini 2020 · M1 · 16 GBRTX 4060 Ti · 16 GB
Qwen3 30B A3B Instruct 2507Mac mini 2024 · M4 · 32 GBRTX 5090 · 32 GB

A dash means no profile of that kind in the catalog has enough memory for the size.

Memory by file precision

The same planning formula at three precisions (0.625, 1.125 and 2 bytes per parameter, plus working room and the system reserve). Estimates, not measurements.

Qwen3 memory estimate by precision
Size
Qwen3 1.7B104115137
Qwen3 4B Instruct 25071151481812
Qwen3 8B15920142822
Qwen3 14B201428224438
Qwen3 30B A3B Instruct 2507312550448276

Family tree

Family, sizes and, for each size, the two 4-bit files the tables use (GGUF Q4_K_M and MLX 4-bit) when one was collected. Other precisions are not drawn here.

Qwen3 family: sizes and published filesSwipe sideways to see the files, or read the list below.
Qwen3 family: sizes and published filesFamilySizesFilesQwen3Alibaba1.7BGGUF Q4_K_M1.1 GBMLX 4-bit1 GB4B Instruct 2507GGUF Q4_K_M2.5 GBMLX 4-bit2.3 GB8BGGUF Q4_K_M5 GBMLX 4-bit4.6 GB14BGGUF Q4_K_M9 GBMLX 4-bit8.3 GB30B A3B Instruct 25073.3B activeGGUF Q4_K_M18.6 GBMLX 4-bit17.2 GB

The same tree as a list:

  • Qwen3 · Alibaba
    • 1.7B
      • GGUF Q4_K_M · 1.1 GB · unsloth/Qwen3-1.7B-GGUF, read Oct 23, 2026
      • MLX 4-bit · 1 GB · lmstudio-community/Qwen3-1.7B-MLX-4bit, read Oct 23, 2026
    • 4B Instruct 2507
      • GGUF Q4_K_M · 2.5 GB · unsloth/Qwen3-4B-Instruct-2507-GGUF, read Oct 23, 2026
      • MLX 4-bit · 2.3 GB · mlx-community/Qwen3-4B-Instruct-2507-4bit, read Oct 23, 2026
    • 8B
      • GGUF Q4_K_M · 5 GB · Qwen/Qwen3-8B-GGUF, read Oct 23, 2026
      • MLX 4-bit · 4.6 GB · mlx-community/Qwen3-8B-4bit, read Oct 23, 2026
    • 14B
      • GGUF Q4_K_M · 9 GB · Qwen/Qwen3-14B-GGUF, read Oct 23, 2026
      • MLX 4-bit · 8.3 GB · lmstudio-community/Qwen3-14B-MLX-4bit, read Oct 23, 2026
    • 30B A3B Instruct 2507 · 3.3B active
      • GGUF Q4_K_M · 18.6 GB · unsloth/Qwen3-30B-A3B-Instruct-2507-GGUF, read Oct 23, 2026
      • MLX 4-bit · 17.2 GB · lmstudio-community/Qwen3-30B-A3B-Instruct-2507-MLX-4bit, read Oct 23, 2026

Learn more

Sources

Where the family and its sizes are named, and where each number comes from.