largeQwen License72B paramsOpen weights

Qwen2.5 72B

72B parameter open-weight model. 72B with 128K context. Q4 requires 45 GB VRAM — multi-GPU (2×RTX 3090/4090) or Mac Studio with 64 GB+ unified memory. Near-frontier open-weight reasoning and multilingual performance.

Alibaba Qwen · Qwen 2.5

Model overview

Qwen2.5 72B is Alibaba's largest widely-deployed Qwen2.5 checkpoint, offering near-frontier reasoning, coding, and multilingual performance among open-weight models. Its size puts it well outside single consumer-GPU territory: Q4 quantization alone needs about 45 GB of VRAM, so realistic local deployment means a dual RTX 3090/4090 setup or a Mac Studio with 64 GB or more of unified memory. It's a strong choice if you already have that hardware and want open-weight quality closer to frontier hosted models, but for a single mid-range GPU, the smaller Qwen2.5 models in the same family are a more practical starting point.

Check your GPU fitWhat is VRAM?What is quantization?Local LLM quickstart

Editorial review

Reviewed byOpenSourcesAI EditorialLast updatedJune 2026SourcesHuggingFace model card (Qwen/Qwen2.5-72B-Instruct), official docs, OpenSourcesAI editorial review.

VRAM figures are empirical estimates. Actual usage varies by runtime, context length, and system configuration. Verify on your specific hardware before production use.

Will Qwen2.5 72B run on your machine?

Qwen2.5 72B is 72B parameters and needs 46.5 GB of VRAM at Q4_K_M — 45 GB of weights plus 1.5 GB of runtime overhead for the inference server itself.

VRAM by quantization

QuantizationWeightsNeeds (with overhead)Quality
Q4_K_M45 GB46.5 GBgood
Q8_080 GB81.5 GBhigh
FP16144 GB145.5 GBreference

Fit on common hardware at Q4_K_M

HardwareMemory the model can useSystem RAMVerdict
CPU OnlyNone (CPU only)16 GBToo large
RTX 4060 Laptop8 GB16 GBToo large
RTX 3060 (12GB)12 GB32 GBToo large
RTX 4060 Ti (16GB)16 GB32 GBToo large
RTX 309024 GB64 GBToo large
Apple Silicon (Unified Memory) 36 GB27 GB of 36 GB36 GBToo large
RTX 509032 GB64 GBToo large
Apple Silicon (Unified Memory) 48 GB36 GB of 48 GB48 GBToo large
Apple Silicon (Unified Memory) 64 GB48 GB of 64 GB64 GBTight
Apple Silicon (Unified Memory) 96 GB72 GB of 96 GB96 GBComfortable
Apple Silicon (Unified Memory) 128 GB96 GB of 128 GB128 GBComfortable
Apple Silicon (Unified Memory) 192 GB144 GB of 192 GB192 GBComfortable

Comfortable means VRAM clears the requirement by 2 GB or more. Tight means it covers the requirement with no margin. CPU offload means the model does not fit in VRAM but system RAM is at least 1.6× the weights, so it will run at reduced speed — expect roughly 1–5 tokens per second. Figures are weights plus a fixed runtime overhead and exclude KV-cache growth, which scales with context length.

Apple Silicon shares one pool of memory between the system and the GPU, so a model cannot use all of it. These rows apply the same 75% usable fraction the Compatibility Checker uses, which is why a 36 GB Mac is graded on less than 36 GB.

Check Qwen2.5 72B against your exact hardwareHow these verdicts are calculated

Need more hardware for Qwen2.5 72B? Open the PC Builder for the 70B tier →

Related models

Llama 3.3 70B70B parameter model — shared focus: chat, coding, rag.Qwen3.6 35B (A3B)36B parameter model — shared focus: chat, reasoning, long-context.Laguna XS 2.133.4B parameter model — shared focus: coding, agents, tool-use.Qwen2.5 14BBuilders hosting local AI on a single RTX 4080 (16 GB) or RTX 4090 (24 GB) who want near-frontier multilingual reasoning, mathematical accuracy, and agentic coding performance without multi-GPU infrastructure. The sweet spot between consumer accessibility and frontier-class open-weight capability for 2026 local deployment.

Ready to run this model locally?

Find a compatible interface in our Local AI Tools directory →