Qwen3 30B quants compared
27 GGUF builds by real file size, probed from bartowski/Qwen_Qwen3-30B-A3B-GGUF on Hugging Face (2026-09-02). 30B params, 3B active.
Download Qwen3 30B Q4_K_M (17.35 GB) — it fits 24 GB of memory with 16k context. With Ollama: ollama run qwen3:30b
Every Qwen3 30B quant by real file size
| Quant | Weights | + KV (16k) | Total | Fits comfortably in | Quality |
|---|---|---|---|---|---|
| BF16 | 56.9 GB | 4.0 GB | 60.9 GB | 96 GB | Full precision (lossless) |
| Q8_0 | 30.25 GB | 4.0 GB | 34.3 GB | 48 GB | Near-lossless |
| Q6_K_L | 23.52 GB | 4.0 GB | 27.5 GB | 32 GB | Excellent |
| Q6_K | 23.38 GB | 4.0 GB | 27.4 GB | 32 GB | Excellent |
| Q5_K_L | 20.43 GB | 4.0 GB | 24.4 GB | 32 GB | Very high |
| Q5_K_M | 20.25 GB | 4.0 GB | 24.3 GB | 32 GB | Very high |
| Q5_K_S | 19.65 GB | 4.0 GB | 23.6 GB | 32 GB | Very high |
| Q4_K_M * | 17.35 GB | 4.0 GB | 21.4 GB | 24 GB | High — the default pick |
| Q4_1 | 17.89 GB | 4.0 GB | 21.9 GB | 32 GB | High |
| Q4_K_L | 17.57 GB | 4.0 GB | 21.6 GB | 24 GB | High |
| Q4_K_S | 16.75 GB | 4.0 GB | 20.8 GB | 24 GB | High |
| Q4_0 | 16.42 GB | 4.0 GB | 20.4 GB | 24 GB | High |
| IQ4_NL | 16.19 GB | 4.0 GB | 20.2 GB | 24 GB | High |
| IQ4_XS | 15.33 GB | 4.0 GB | 19.3 GB | 24 GB | High |
| Q3_K_XL | 13.84 GB | 4.0 GB | 17.8 GB | 24 GB | Acceptable — visible loss |
| Q3_K_L | 13.58 GB | 4.0 GB | 17.6 GB | 24 GB | Acceptable — visible loss |
| IQ3_M | 13.11 GB | 4.0 GB | 17.1 GB | 24 GB | Acceptable — visible loss |
| Q3_K_M | 13.11 GB | 4.0 GB | 17.1 GB | 24 GB | Acceptable — visible loss |
| Q3_K_S | 12.51 GB | 4.0 GB | 16.5 GB | 24 GB | Acceptable — visible loss |
| IQ3_XS | 11.86 GB | 4.0 GB | 15.9 GB | 24 GB | Acceptable — visible loss |
| IQ3_XXS | 11.38 GB | 4.0 GB | 15.4 GB | 24 GB | Acceptable — visible loss |
| Q2_K_L | 10.44 GB | 4.0 GB | 14.4 GB | 24 GB | Experimental — not ranked — never recommended |
| Q2_K | 10.16 GB | 4.0 GB | 14.2 GB | 16 GB | Experimental — not ranked — never recommended |
| IQ2_M | 9.71 GB | 4.0 GB | 13.7 GB | 16 GB | Experimental — not ranked — never recommended |
| IQ2_S | 8.59 GB | 4.0 GB | 12.6 GB | 16 GB | Experimental — not ranked — never recommended |
| IQ2_XS | 8.51 GB | 4.0 GB | 12.5 GB | 16 GB | Experimental — not ranked — never recommended |
| IQ2_XXS | 7.59 GB | 4.0 GB | 11.6 GB | 16 GB | Experimental — not ranked — never recommended |
* default pick. Weights = real GGUF file sizes from bartowski/Qwen_Qwen3-30B-A3B-GGUF (probed 2026-09-02). KV = fp16 estimate; a q8_0 cache roughly halves it. "Comfortable" = weights + KV within 90% of memory.
Best Qwen3 30B quant by memory
| Memory | Recommended quant | Total (16k ctx) |
|---|---|---|
| 24 GB | Q4_K_M | 21.4 GB |
| 32 GB | Q6_K | 27.4 GB |
| 48 GB | Q8_0 | 34.3 GB |
| 96 GB | BF16 | 60.9 GB |
Why we don't rank Qwen3 30B's 2-bit quants
Quants at 2 bits per weight or below (Q2_K, IQ2, IQ1, TQ1) cut file size by roughly half versus Q4, but the quality collapse is steep and non-linear: perplexity spikes, instruction-following degrades, and hallucinations rise. A model that answers faster but wrong is not a smaller model — it is a worse one. ModelFit lists these builds for completeness but never ranks or recommends them.
Frequently asked questions
What is the best quantization of Qwen3 30B?
Q4_K_M is the default pick: 17.35 GB of weights, high — the default pick quality, fitting comfortably in 24 GB of memory (weights + 16k context KV-cache). Go Q6_K or Q8_0 if you have headroom.
How much memory does Qwen3 30B need?
At Q4_K_M, Qwen3 30B needs 17.35 GB for the weights plus ~4.0 GB of KV-cache at 16k context — about 21.4 GB total, so a 24 GB card or Mac (90% usable budget) runs it comfortably.
Should I use a Q2_K or IQ2 quant of Qwen3 30B?
No. Qwen3 30B at 2 bits per weight is a visibly worse model — quality collapse at that bitrate is steep, not gradual. If only a 2-bit build fits your memory, run a smaller model at Q4_K_M instead. ModelFit lists these builds but never recommends them.
Cite this page
ModelFit: Qwen3 30B quantization comparison (real GGUF file sizes). https://modelfit.io/quant-compare/qwen3-30b-a3b/ (data probed 2026-09-02, CC BY 4.0).