Qwen3.8 VRAM requirements
Memory needed to run Qwen3.8 at every published size and quant, with an 8k context. Each figure is weights plus KV cache plus runtime buffers, from the same memory model as the VRAM calculator.
Total memory by size and quant
GB at an 8k context. Bold is the Q4_K_M column, the usual starting point.
| Size | Q8_0 | Q6_K | Q5_K_M | Q4_K_M | Q3_K_M |
|---|---|---|---|---|---|
| 27B | 29 | 24 | 21 | 17 | 14 |
| 35B | 37 | 31 | 26 | 22 | 18 |
| 122B · 10B active | 125 | 102 | 86 | 71 | 56 |
| 125B · 6B active | 127 | 104 | 88 | 72 | 57 |
| 177B | 179 | 146 | 124 | 102 | 80 |
| 2400B · 95B active | 2402 | 1952 | 1652 | 1352 | 1052 |
KV cache by context length
GB the context alone adds on top of the weights. Add this to a weights figure to size a longer session.
| Size | Weights at Q4_K_M | 4k ctx | 16k ctx | 32k ctx | 128k ctx |
|---|---|---|---|---|---|
| 27B | 15 | 0.3 | 1.1 | 2.1 | 8.6 |
| 35B | 20 | 0.3 | 1.1 | 2.1 | 8.6 |
| 122B · 10B active | 69 | 0.4 | 1.5 | 3.0 | 12 |
| 125B · 6B active | 70 | 0.1 | 0.4 | 0.8 | 3.2 |
| 177B | 100 | 0.3 | 1.1 | 2.1 | 8.6 |
| 2400B · 95B active | 1350 | 0.3 | 1.1 | 2.1 | 8.6 |
What each size fits on at Q4_K_M
Hardware whose memory holds the model with headroom, smallest pool first. Fits means under 85% of the pool at an 8k context. A Mac's pool is what macOS lets the GPU use - two thirds of its memory up to 32 GB, three quarters from 36 GB - not the whole of it.
- 27Bneeds about 17 GB
Smallest: NVIDIA RTX 2080 Ti 22GB (modded) (22 GB), AMD RX 7900 XTX (24 GB), Intel Arc Pro B60 24GB (24 GB), NVIDIA A10G (24 GB), and 57 more.
On a Mac: 32 GB or more, with macOS's default GPU limit.
- 35Bneeds about 22 GB
Smallest: AMD MI50 32GB (32 GB), AMD Radeon AI PRO R9700 32GB (32 GB), AMD Radeon Pro V620 32GB (32 GB), AMD Radeon Pro W6800 (32 GB), and 41 more.
On a Mac: 36 GB or more, with macOS's default GPU limit.
- 122B · 10B activeneeds about 71 GB
Smallest: NVIDIA H100 NVL (94 GB), NVIDIA RTX Pro 6000 Blackwell (96 GB), NVIDIA RTX PRO 6000 Max-Q (96 GB), AMD Instinct MI250 (128 GB), and 13 more.
On a Mac: 128 GB or more, with macOS's default GPU limit.
- 125B · 6B activeneeds about 72 GB
Smallest: NVIDIA H100 NVL (94 GB), NVIDIA RTX Pro 6000 Blackwell (96 GB), NVIDIA RTX PRO 6000 Max-Q (96 GB), AMD Instinct MI250 (128 GB), and 13 more.
On a Mac: 128 GB or more, with macOS's default GPU limit.
- 177Bneeds about 102 GB
Smallest: AMD Instinct MI250 (128 GB), AMD Instinct MI250X 128GB (128 GB), AMD Instinct MI300A (128 GB), AMD Strix Halo 128GB (128 GB), and 10 more.
On a Mac: 192 GB or more, with macOS's default GPU limit.
- 2400B · 95B activeneeds about 1352 GB
Does not fit any single GPU or Mac on llamaperf at Q4_K_M. It needs a multi-GPU rig, a smaller quant, or CPU offloading.
Frequently asked
How much VRAM does Qwen3.8 need?
From about 17 GB for the 27B model to about 1352 GB for the 2400B · 95B active model, at Q4_K_M with an 8k context and including the KV cache and runtime buffers. Q8 needs more and Q3 less; the table above lists every size and rung.
Can I run Qwen3.8 on a 16GB GPU?
Not comfortably. Even the 27B model needs about 17 GB at Q4_K_M with an 8k context, which is more than a 16 GB card holds with headroom. A lower quant or CPU offloading can still get it running, at a cost in quality or speed.
Can I run Qwen3.8 on a 24GB GPU such as an RTX 3090 or 4090?
Yes. The 27B model needs about 17 GB at Q4_K_M with an 8k context, which fits a 24 GB card with headroom. Larger sizes need a smaller quant or more memory.
How are these numbers calculated?
Weights at the quant's bits per weight, plus a KV cache sized from the model's attention design for the chosen context, plus compute buffers and a fixed runtime allowance. It is the same memory model the VRAM calculator uses. A family whose attention profile is not on file is sized as plain grouped-query attention, which errs towards needing more memory.