QwQ 32B Preview — Hardware Requirements & GPU Compatibility
ChatReasoning
8.4K downloads
1.7K likesSpecifications
- Publisher
- Alibaba
- Family
- QwQ
- Parameters
- 32.8B
- Architecture
- Qwen2ForCausalLM
- Context Length
- 32,768 tokens
- Vocabulary Size
- 152,064
- Release Date
- 2025-01-12
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does QwQ 32B Preview Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_K | 3.40 | 14.8 GB | 22.8 GB | 13.92 GB | 2-bit quantization with K-quant improvements |
| Q3_K_M | 3.90 | 16.8 GB | 24.9 GB | 15.97 GB | 3-bit medium quantization |
| Q3_K_L | 4.10 | 17.6 GB | 25.7 GB | 16.79 GB | 3-bit large quantization |
| Q4_K_M | 4.80 | 20.5 GB | 28.6 GB | 19.66 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_M | 5.70 | 24.2 GB | 32.2 GB | 23.34 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_K | 6.60 | 27.9 GB | 35.9 GB | 27.03 GB | 6-bit quantization, very good quality |
| Q8_0 | 8.00 | 33.6 GB | 41.6 GB | 32.76 GB | 8-bit quantization, near-lossless |
Which GPUs Can Run QwQ 32B Preview?
Q4_K_M · 20.5 GBQwQ 32B Preview (Q4_K_M) requires 20.5 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 27+ GB is recommended. Using the full 33K context window can add up to 8.1 GB, bringing total usage to 28.6 GB. 5 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti.
Runs great
— Plenty of headroom
Decent
— Enough VRAM, may be tightWhich Devices Can Run QwQ 32B Preview?
Q4_K_M · 20.5 GB21 devices with unified memory can run QwQ 32B Preview, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, Mac Mini M4 Pro (24 GB).
Runs great
— Plenty of headroomRelated Models
Derivatives (4)
Frequently Asked Questions
- How much VRAM does QwQ 32B Preview need?
QwQ 32B Preview requires 20.5 GB of VRAM at Q4_K_M, or 33.6 GB at Q8_0. Full 33K context adds up to 8.1 GB (28.6 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 32.8B × 4.8 bits ÷ 8 = 19.7 GB
KV Cache + Overhead ≈ 0.8 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 8.9 GB (at full 33K context)
VRAM usage by quantization
Q4_K_M20.5 GBQ4_K_M + full context28.6 GB- Can NVIDIA GeForce RTX 4090 run QwQ 32B Preview?
Yes, at Q4_K_M (20.5 GB) or lower. Higher quantizations like Q5_K_M (24.2 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
- What's the best quantization for QwQ 32B Preview?
For QwQ 32B Preview, Q4_K_M (20.5 GB) offers the best balance of quality and VRAM usage. Q5_K_M (24.2 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 14.8 GB.
VRAM requirement by quantization
Q2_K14.8 GB~75%Q3_K_L17.6 GB~86%Q4_K_M ★20.5 GB~89%Q5_K_M24.2 GB~92%Q6_K27.9 GB~95%Q8_033.6 GB~99%★ Recommended — best balance of quality and VRAM usage.
- Can I run QwQ 32B Preview on a Mac?
QwQ 32B Preview requires at least 14.8 GB at Q2_K, 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 QwQ 32B Preview locally?
Yes — QwQ 32B Preview can run locally on consumer hardware. At Q4_K_M quantization it needs 20.5 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is QwQ 32B Preview?
At Q4_K_M, QwQ 32B Preview can reach ~142 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~32 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 ÷ 20.5 × 0.55 = ~142 tok/s
Estimated speed at Q4_K_M (20.5 GB)
AMD Instinct MI300X~142 tok/sNVIDIA GeForce RTX 4090~32 tok/sNVIDIA H100 SXM~106 tok/sAMD Instinct MI250X~88 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of QwQ 32B Preview?
At Q4_K_M, the download is about 19.66 GB. The full-precision Q8_0 version is 32.76 GB. The smallest option (Q2_K) is 13.92 GB.
- Which GPUs can run QwQ 32B Preview?
5 consumer GPUs can run QwQ 32B Preview at Q4_K_M (20.5 GB). Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090. 1 GPU have plenty of headroom for comfortable inference.
- Which devices can run QwQ 32B Preview?
21 devices with unified memory can run QwQ 32B Preview at Q4_K_M (20.5 GB), including Mac Mini M4 (32 GB), Mac Mini M4 Pro (24 GB), Mac Mini M4 Pro (48 GB), Mac Pro M2 Ultra (192 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.