Best local AI models for NVIDIA 930A

The largest models that fit fully

The 30 largest of the 56 models that fit; every smaller model in the catalog fits too. Best quant means the highest quality compression whose weights and 4k context both sit inside the memory.

Close, but only with CPU offload

These need more than the card holds at their smallest practical quant, so part of the model runs from system memory (figures assume 32 GB of it). They work, several times slower.

How to read this

The NVIDIA 930A is an entry level graphics card equipped with 2 GB of DDR3 memory. This dedicated video memory determines the size of the artificial intelligence models you can run locally. Because the hardware has a strict 2 GB limit, selecting the correct model size and quantization level is necessary to avoid out of memory errors during inference.

Quantization is a compression method that reduces the precision of model weights. The quant column shows the best format to fit each model into the available video memory. For example, the Allegro 2.8B model fits into 2 GB of video memory when using the Q4_K_M quantization. Smaller models like the SmolVLM 2B can run at a higher quality Q6_K quantization while still staying within the 2 GB limit.

Running models at or near the 2 GB limit introduces a strict context window caveat. If you use a 4k context window, the memory required for the context history will exceed the remaining video memory. For models like the LFM2 2.6B or the Stable Diffusion 3 Medium, you must reduce the context length to prevent the system from slowing down or crashing.

When a model is too large for the 2 GB video memory, you can use CPU offload. This technique splits the model weights between the graphics card and your system RAM. To use CPU offload, this guide assumes your computer has 32 GB of system RAM. Offloading allows you to run larger models but it reduces the generation speed because the system RAM is slower than the graphics card memory.

Using CPU offload enables you to run models like the SmolLM3 3B or the Replit Code v1.5 3B. These models need 2.2 GB of video memory at Q4_K_M quantization and require 4.2 GB of system RAM. You can also run image generation models like Stable Diffusion XL which requires 4.1 GB at FP8 or optimized settings along with 6.1 GB of system RAM.