Best local AI models for NVIDIA GTX 950A

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 GTX 950A is an entry level graphics hardware option equipped with 2 GB of DDR3 video memory. This memory size determines the maximum size of the artificial intelligence models you can run locally. Because the onboard memory is limited to 2 GB, choosing the correct model size and quantization level is critical to prevent out of memory errors during inference.

Quantization is a method that compresses model weights to save video memory. In our catalog, the best quant column shows the optimal compression level for each model. For example, the Allegro 2.8B model fits in 2 GB of video memory when compressed to the Q4_K_M quantization. Smaller models like Moondream 2 can run at the higher quality Q6_K quantization while using 1.9 GB of video memory.

Running models at their limit leaves very little room for context. If you load a 2B model like Stable Diffusion 3 Medium at Q6_K using 2 GB of video memory, you will experience severe speed penalties if your prompt or context window expands. For text generation, keeping a buffer for a 4k context window is highly recommended to maintain acceptable generation speeds.

When a model exceeds the 2 GB physical limit of your graphics card, you must use CPU offload. This process splits the model weights between your video memory and your system RAM. For this setup, we assume your computer has 32 GB of system RAM available to handle the overflow.

Offloading comes with a performance cost. Loading the SmolLM3 3B model at Q4_K_M requires 2.2 GB of video memory and 4.2 GB of system RAM, which slows down processing because data must travel over the system bus. Similarly, running Stable Diffusion XL at FP8 requires 4.1 GB of video memory and 6.1 GB of system RAM, resulting in much longer generation times compared to running fully inside the graphics memory.