What can you run on a 32GB graphics card?

With 32GB VRAM you can run 14 of the 30 open-source models we track — up to about 36B parameters at Q4 quantization. The most popular that fits is Qwen3.5-9B, needing 6.8 GB and generating around ~205 tok/s on typical hardware of this size.

Typical machines: RTX 5090. After the operating system takes its share, about 33 GB is free for a model.

Having the memory is not the same as having the speed

32GB VRAM decides whether a model loads. How fast it answers is decided by memory bandwidth — how quickly your machine can read the model, which it must do once for every word it writes. The table quotes a RTX 5090; pick yours below to see the difference. Why bandwidth and not the processor →

1792 GB/sGeForce RTX 5090 (32GB)· NVIDIA published specification: 1,792 GB/s GDDR7, 512-bit

Models that run on 32GB VRAM in 2026

ModelSizeMemoryEst. speedFeels like
Qwen3.5-9BAlibaba / Qwen9.7B6.8 GBRuns well~205 tok/sFast
gpt-oss-20bOpenAI · MoE22B13 GBRuns well~501 tok/s4.3B activeFast
Qwen3.5-4BAlibaba / Qwen4.7B3.8 GBRuns well~425 tok/sFast
Qwen3.5-0.8BAlibaba / Qwen873M1.3 GBRuns well~2187 tok/sFast
Qwen3.5-27BAlibaba / Qwen28B19 GBRuns well~70 tok/sFast
Qwen3.5-35B-A3BAlibaba / Qwen · MoE36B25 GBRuns well~265 tok/s7.2B activeFast
Qwen3.5-2BAlibaba / Qwen2.3B2.0 GBRuns well~909 tok/sFast
GLM-4.7-FlashZhipu AI31B22 GBRuns well~62 tok/sFast
Olmo-3-7B-InstructAi27.3B7.2 GBRuns well~260 tok/sFast
GLM-4.6V-FlashZhipu AI10B7.0 GBRuns well~189 tok/sFast
ERNIE-4.5-VL-28B-A3B-PTBaidu AI · MoE29B20 GBRuns well~66 tok/sFast
Olmo-3.1-32B-InstructAi232B23 GBRuns well~60 tok/sFast
ERNIE-4.5-21B-A3B-PTBaidu AI · MoE22B15 GBRuns well~86 tok/sFast
ERNIE-4.5-0.3B-PTBaidu AI361M853 MBRuns well~4840 tok/sFast

All figures at Q4_K_M quantization, 4K context. “Memory” includes the model, its context cache and runtime overhead. Speed is calculated, not benchmarked: 1792 GB/s × 0.65 ÷ bytes read per token, assuming a RTX 5090. Scale it by your own machine's bandwidth from the list above. Mixture-of-experts models read only their active parameters per token, which is why some large ones outrun smaller dense models.

Image generators that also fit

These make pictures rather than text. Their memory use scales with the resolution you generate at, not conversation length, so treat these figures as a floor rather than a ceiling.

What you'd need more memory for

The next models up, and what they ask for.

Other hardware

Verified against HuggingFace on 2026-08-01New to this? Start here →Runs on a laptop and up · last 12 months only