How much VRAM does Qwen3-72B-Synthesis need?

Q4_K_M is exactly 47,369,123,104 bytes (44.12 GiB) in weights. Add the KV cache, which depends on context length, plus roughly half a gigabyte of runtime overhead.

How large is Qwen3-72B-Synthesis's KV cache?

10.00 GiB at 32K context with an f16 cache, computed per layer from the model's architecture. Quantizing the cache to q8_0 roughly halves it.

QuixiAI · text

Qwen3-72B-Synthesis

QuixiAI/Qwen3-72B-Synthesis

Qwen3-72B-Synthesis at Q4_K_M is exactly 47,369,123,104 bytes (44.12 GiB / 47.37 GB) — an effective 5.213 bits per weight, not the nominal 4. Its KV cache at 32K is 10.00 GiB.

From the file· summed from 1 file(s)From the file· KV per layer

Parameters

72.7B

Architecture

qwen3

80 layers

Context

40,960

native (config.json)

License

apache-2.0

Shipped quantizations

● exact bytes, summed from published files

Quant	Size●	Exact bytes●	Effective bpw●	Tensors●	Publisher
Q2_K	27.68 GiB	29,723,473,184	3.271	—	mradermacher
Q3_K_S	31.95 GiB	34,310,264,096	3.776	—	mradermacher
Q3_K_M	35.03 GiB	37,612,426,528	4.139	—	mradermacher
Q3_K_L	36.79 GiB	39,500,715,296	4.347	—	mradermacher
IQ4_XS	37.40 GiB	40,158,623,008	4.419	—	mradermacher
Q4_K_S	40.80 GiB	43,804,882,208	4.820	—	mradermacher
Q4_K_M	44.12 GiB	47,369,123,104	5.213	—	mradermacher

KV cache by context

computed per layer

Context	KV cache (f16)●	Flat formula	Overstated by	Full / windowed / recurrent
4,096	1.25 GiB	1.25 GiB	—	80 / 0 / 0
8,192	2.50 GiB	2.50 GiB	—	80 / 0 / 0
16,384	5.00 GiB	5.00 GiB	—	80 / 0 / 0
32,768	10.00 GiB	10.00 GiB	—	80 / 0 / 0
65,536	20.00 GiB	20.00 GiB	—	80 / 0 / 0
131,072	40.00 GiB	40.00 GiB	—	80 / 0 / 0

Compare with

same modality, comparable size

vs Qwen3-Coder-30B-A3B-Instruct vs Qwen3.6-27B vs Qwen3.8-27B vs gemma-4-E4B-it vs Qwen3-30B-A3B-Thinking-2507 vs Qwen3-8B All alternatives →Will it fit on my card? →

Why other calculators give a different number

A parameters × bits ÷ 8 estimate puts Q4_K_M at roughly 38.08 GiB. The real file is 44.12 GiB, because a quantization is a mixture and some tensors are always kept at higher precision.

Architecture

from config.json

Layers

Attention heads

KV heads

Head dim

128

Hidden size

8192

Vocab

151,669

Sliding window

none

SWA period

—

MLA

Experts

—

Experts per token

—

use_sliding_window

false

Questions people ask

How much VRAM does Qwen3-72B-Synthesis need?: Q4_K_M is exactly 47,369,123,104 bytes (44.12 GiB) in weights. Add the KV cache, which depends on your context length, plus roughly half a gigabyte of runtime overhead.
How large is Qwen3-72B-Synthesis's KV cache?: 10.00 GiB at 32K context with an f16 cache, computed per layer. Quantizing the cache to q8_0 roughly halves it, which is often the difference between a context length fitting and not.
Which quantization of Qwen3-72B-Synthesis should I use?: Q4_K_M is the usual default. Pick the largest quantization that fits your card at the context you actually need — the table above gives exact sizes for every one published.