Qwen3 4B Domino B16 — Hardware Requirements & GPU Compatibility
Chat
Qwen3 4B Domino B16 is a 588M-parameter open language model from Huang2020 in the Qwen family. It supports a context window of up to 40,960 tokens. At Q4_K_M it needs about 0.68 GB of VRAM — see which GPUs and Macs can run it below.
365 downloads
2 likesSpecifications
- Publisher
- Huang2020
- Family
- Qwen
- Parameters
- 588M
- Architecture
- DFlashDraftModel
- Context Length
- 40,960 tokens
- Vocabulary Size
- 151,936
- Release Date
- 2026-06-01
Get Started
HuggingFace
How Much VRAM Does Qwen3 4B Domino B16 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q3_K_S | 3.50 | 0.6 GB | 1.1 GB | 0.26 GB | 3-bit small quantization |
| Q2_K | 3.40 | 0.6 GB | 1.1 GB | 0.25 GB | 2-bit quantization with K-quant improvements |
| Q3_K_M | 3.90 | 0.6 GB | 1.1 GB | 0.29 GB | 3-bit medium quantization |
| Q4_0 | 4.00 | 0.6 GB | 1.1 GB | 0.29 GB | 4-bit legacy quantization |
| Q4_K_M | 4.80 | 0.7 GB | 1.2 GB | 0.35 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 0.8 GB | 1.2 GB | 0.42 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 0.8 GB | 1.3 GB | 0.49 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 0.9 GB | 1.4 GB | 0.59 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run Qwen3 4B Domino B16?
Q4_K_M · 0.7 GBQwen3 4B Domino B16 (Q4_K_M) requires 0.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 1+ GB is recommended. Using the full 41K context window can add up to 0.5 GB, bringing total usage to 1.2 GB. 35 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroomWhich Devices Can Run Qwen3 4B Domino B16?
Q4_K_M · 0.7 GB33 devices with unified memory can run Qwen3 4B Domino B16, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does Qwen3 4B Domino B16 need?
Qwen3 4B Domino B16 requires 0.7 GB of VRAM at Q4_K_M, or 0.9 GB at Q8_0.
VRAM = Weights + KV Cache + Overhead
Weights = 588M × 4.8 bits ÷ 8 = 0.4 GB
KV Cache + Overhead ≈ 0.3 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 0.8 GB (at full 41K context)
VRAM usage by quantization
Q4_K_M0.7 GBQ4_K_M + full context1.2 GB- What's the best quantization for Qwen3 4B Domino B16?
For Qwen3 4B Domino B16, Q4_K_M (0.7 GB) offers the best balance of quality and VRAM usage. Q5_0 (0.7 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 0.5 GB.
VRAM requirement by quantization
IQ2_XXS0.5 GBQ2_K0.6 GBIQ4_NL0.7 GBQ4_K_M ★0.7 GBQ5_00.7 GBQ8_00.9 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run Qwen3 4B Domino B16 on a Mac?
Qwen3 4B Domino B16 requires at least 0.5 GB at IQ2_XXS, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run Qwen3 4B Domino B16 locally?
Yes — Qwen3 4B Domino B16 can run locally on consumer hardware. At Q4_K_M quantization it needs 0.7 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is Qwen3 4B Domino B16?
At Q4_K_M, Qwen3 4B Domino B16 can reach ~4287 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~964 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 0.7 × 0.55 = ~4287 tok/s
Estimated speed at Q4_K_M (0.7 GB)
~4287 tok/s~964 tok/s~3204 tok/s~2650 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of Qwen3 4B Domino B16?
At Q4_K_M, the download is about 0.35 GB. The full-precision Q8_0 version is 0.59 GB. The smallest option (IQ2_XXS) is 0.16 GB.
- Which GPUs can run Qwen3 4B Domino B16?
35 consumer GPUs can run Qwen3 4B Domino B16 at Q4_K_M (0.7 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 35 GPUs have plenty of headroom for comfortable inference.
- Which devices can run Qwen3 4B Domino B16?
33 devices with unified memory can run Qwen3 4B Domino B16 at Q4_K_M (0.7 GB), including Mac Mini M4 (16 GB), Mac Mini M4 (32 GB), Mac Mini M4 Pro (24 GB), Mac Mini M4 Pro (48 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.