Skip to content

Qwen2 1.5B VRAM requirements

How much VRAM Qwen2 1.5B needs
4-bit weights: 893 MBWith 8K context: 1.9 GBWith 32K context: 2.5 GB8-bit, 8K context: 2.5 GB
Parameters
1.5B
Longest context
128K
Released
May 2024
License
Apache 2.0
Strongest at
Not rated

How much VRAM Qwen2 1.5B needs

Memory Qwen2 1.5B needs at each precision and context length
PrecisionWeightsWith 8KWith 32KWith 128K
16-bit2.9 GB3.9 GB4.5 GB7.2 GB
8-bit1.5 GB2.5 GB3.2 GB5.8 GB
6-bit1.2 GB2.2 GB2.8 GB5.5 GB
5-bit1.0 GB2.0 GB2.7 GB5.3 GB
4-bit893 MB1.9 GB2.5 GB5.2 GB
3-bit720 MB1.7 GB2.4 GB5.0 GB
2-bit552 MB1.5 GB2.2 GB4.8 GB
1.6-bit304 MB1.3 GB2.0 GB4.6 GB

One conversation, 16-bit context cache, runtime overhead included.

GPUs and Macs that run Qwen2 1.5B

The best way Qwen2 1.5B runs on common GPUs, Macs and AI boxes
MachineBest fitContextSpeedFirst replyReport
RTX 40608-bit128K106 tok/s0.2 sPDF: Qwen2 1.5B on RTX 4060 report
RTX 50608-bit128K165 tok/s0.2 sPDF: Qwen2 1.5B on RTX 5060 report
RTX 30708-bit128K165 tok/s0.1 sPDF: Qwen2 1.5B on RTX 3070 report
RTX 3080 10 GB8-bit128K255 tok/s0.1 sPDF: Qwen2 1.5B on RTX 3080 10 GB report
RTX 3060 12 GB8-bit128K136 tok/s0.2 sPDF: Qwen2 1.5B on RTX 3060 12 GB report
Arc B5808-bit128K113 tok/s0.2 sPDF: Qwen2 1.5B on Arc B580 report
RTX 40708-bit128K182 tok/s0.1 sPDF: Qwen2 1.5B on RTX 4070 report
RTX 4070 Super8-bit128K182 tok/s0.1 sPDF: Qwen2 1.5B on RTX 4070 Super report
RTX 50708-bit128K231 tok/s0.1 sPDF: Qwen2 1.5B on RTX 5070 report
RTX 4060 Ti 16 GB8-bit128K111 tok/s0.1 sPDF: Qwen2 1.5B on RTX 4060 Ti 16 GB report
RTX 5060 Ti 16 GB8-bit128K165 tok/s0.1 sPDF: Qwen2 1.5B on RTX 5060 Ti 16 GB report
Arc A770 16 GB8-bit128K133 tok/s0.1 sPDF: Qwen2 1.5B on Arc A770 16 GB report
Radeon RX 9070 XT8-bit128K176 tok/s0.1 sPDF: Qwen2 1.5B on Radeon RX 9070 XT report
RTX 4070 Ti Super8-bit128K231 tok/s0.1 sPDF: Qwen2 1.5B on RTX 4070 Ti Super report
RTX 4080 Super8-bit128K249 tok/s0.1 sPDF: Qwen2 1.5B on RTX 4080 Super report
RTX 5070 Ti8-bit128K289 tok/s0.1 sPDF: Qwen2 1.5B on RTX 5070 Ti report
RTX 50808-bit128K305 tok/s0.1 sPDF: Qwen2 1.5B on RTX 5080 report
Radeon RX 7900 XTX8-bit128K235 tok/s0.1 sPDF: Qwen2 1.5B on Radeon RX 7900 XTX report
RTX 30908-bit128K299 tok/s0.1 sPDF: Qwen2 1.5B on RTX 3090 report
RTX 40908-bit128K316 tok/s0.1 sPDF: Qwen2 1.5B on RTX 4090 report
Radeon AI PRO R97008-bit128K176 tok/s0.1 sPDF: Qwen2 1.5B on Radeon AI PRO R9700 report
RTX 50908-bit128K466 tok/s0.1 sPDF: Qwen2 1.5B on RTX 5090 report
2× RTX 30908-bit128K299 tok/s0.1 sPDF: Qwen2 1.5B on 2× RTX 3090 report
2× RTX 40908-bit128K316 tok/s0.1 sPDF: Qwen2 1.5B on 2× RTX 4090 report
RTX PRO 5000 Blackwell 48 GB8-bit128K387 tok/s0.1 sPDF: Qwen2 1.5B on RTX PRO 5000 Blackwell 48 GB report
RTX 6000 Ada8-bit128K305 tok/s0.1 sPDF: Qwen2 1.5B on RTX 6000 Ada report
L40S8-bit128K282 tok/s0.1 sPDF: Qwen2 1.5B on L40S report
2× RTX 50908-bit128K466 tok/s0.1 sPDF: Qwen2 1.5B on 2× RTX 5090 report
4× RTX 30908-bit128K299 tok/s0.1 sPDF: Qwen2 1.5B on 4× RTX 3090 report
RTX PRO 6000 Blackwell8-bit128K466 tok/s0.1 sPDF: Qwen2 1.5B on RTX PRO 6000 Blackwell report
Mac mini M4 16 GB8-bit128K53 tok/s1.9 sPDF: Qwen2 1.5B on Mac mini M4 16 GB report
Mac mini M4 24 GB8-bit128K53 tok/s1.9 sPDF: Qwen2 1.5B on Mac mini M4 24 GB report
Mac mini M4 32 GB8-bit128K53 tok/s1.9 sPDF: Qwen2 1.5B on Mac mini M4 32 GB report
Mac mini M4 Pro 24 GB8-bit128K105 tok/s1.0 sPDF: Qwen2 1.5B on Mac mini M4 Pro 24 GB report
Mac mini M4 Pro 48 GB8-bit128K105 tok/s1.0 sPDF: Qwen2 1.5B on Mac mini M4 Pro 48 GB report
Mac mini M4 Pro 64 GB8-bit128K105 tok/s1.0 sPDF: Qwen2 1.5B on Mac mini M4 Pro 64 GB report
M4 Max 128 GB8-bit128K160 tok/s0.5 sPDF: Qwen2 1.5B on M4 Max 128 GB report
M5 Max 128 GB8-bit128K172 tok/s0.1 sPDF: Qwen2 1.5B on M5 Max 128 GB report
M3 Ultra 96 GB8-bit128K188 tok/s0.3 sPDF: Qwen2 1.5B on M3 Ultra 96 GB report
M3 Ultra 256 GB8-bit128K188 tok/s0.3 sPDF: Qwen2 1.5B on M3 Ultra 256 GB report
M3 Ultra 512 GB8-bit128K188 tok/s0.3 sPDF: Qwen2 1.5B on M3 Ultra 512 GB report
M5 Ultra 512 GB8-bit128K230 tok/s0.1 sPDF: Qwen2 1.5B on M5 Ultra 512 GB report
Ryzen AI Max+ 395 64 GB8-bit128K83 tok/s0.3 sPDF: Qwen2 1.5B on Ryzen AI Max+ 395 64 GB report
Ryzen AI Max+ 395 128 GB8-bit128K83 tok/s0.3 sPDF: Qwen2 1.5B on Ryzen AI Max+ 395 128 GB report
RTX Spark8-bit128K116 tok/s0.1 sPDF: Qwen2 1.5B on RTX Spark report
DGX Spark8-bit128K106 tok/s0.1 sPDF: Qwen2 1.5B on DGX Spark report
Dual DGX Spark8-bit128K106 tok/s0.1 sPDF: Qwen2 1.5B on Dual DGX Spark report
DGX Station GB3008-bit128K855 tok/s0.1 sPDF: Qwen2 1.5B on DGX Station GB300 report
2× RTX PRO 6000 Blackwell8-bit128K466 tok/s0.1 sPDF: Qwen2 1.5B on 2× RTX PRO 6000 Blackwell report
4× RTX PRO 6000 Blackwell8-bit128K466 tok/s0.1 sPDF: Qwen2 1.5B on 4× RTX PRO 6000 Blackwell report
8× RTX PRO 6000 Blackwell8-bit128K466 tok/s0.1 sPDF: Qwen2 1.5B on 8× RTX PRO 6000 Blackwell report
A100 80 GB SXM8-bit128K503 tok/s0.1 sPDF: Qwen2 1.5B on A100 80 GB SXM report
H100 SXM8-bit128K650 tok/s0.1 sPDF: Qwen2 1.5B on H100 SXM report
4× H200 NVL8-bit128K753 tok/s0.1 sPDF: Qwen2 1.5B on 4× H200 NVL report
8× H200 NVL8-bit128K753 tok/s0.1 sPDF: Qwen2 1.5B on 8× H200 NVL report

Best fit is the highest quality, up to 8-bit, that leaves 16K of context. First reply is the wait for a 2K prompt. Speeds are rough, plus or minus 25%.

How to run it

llama.cpp
llama-server -hf legraphista/Qwen2-1.5B-IMat-GGUF
vLLM
vllm serve Qwen/Qwen2-1.5B

Questions

How much VRAM does Qwen2 1.5B need?

Qwen2 1.5B needs 893 MB for its weights at 4-bit, 1.9 GB with an 8K context and 2.5 GB with 32K. At 8-bit, 2.5 GB.

What is the minimum VRAM for Qwen2 1.5B?

About 1.7 GB at 3-bit with an 8K context, the lowest precision we recommend for a model this size. The smallest machine here that runs it without spilling is the RTX 4060.

How big is Qwen2 1.5B?

1.5B parameters: 2.9 GB of weights at 16-bit, 893 MB at 4-bit.

Can Qwen2 1.5B run on an RTX 4090?

Yes. At 8-bit it holds 128K of context and writes about 316 tokens a second.

More models