Best local AI models for NVIDIA 830M

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 830M is an entry level graphics card equipped with 2 GB of DDR3 video memory. This memory size is the absolute limit for running local AI models entirely on your graphics hardware. To fit models inside this tight space you must use quantized versions. Quantization reduces the precision of model weights to save memory. The quant column shows the best format that fits within your hardware limits without causing out of memory errors.

For models under 2 GB you can achieve complete GPU execution. The largest fitting models include Allegro 2.8B and Open-Sora Plan 2.7B which both use 2 GB of video memory at the Q4_K_M quant. Other options like LFM2 2.6B and Playground v2.5 2.6B require 1.9 GB of video memory at Q4_K_M. You can also run Stable Diffusion 3.5 Medium 2.5B and Canary 2.5B using 1.8 GB of video memory at the Q4_K_M quant.

Slightly smaller models can use higher precision quants for better output quality. SeamlessM4T v2 2.3B and Parler-TTS 2.2B run well at the Q5_K_M quant using 2 GB and 1.9 GB of video memory. Kimi K3 DSpark 2.2B also fits at Q5_K_M using 2 GB. Models like SmolVLM 2B, Stable Diffusion 3 Medium 2B, Pyramid Flow 2B, and Wav2Vec2 2B can utilize the Q6_K quant while using exactly 2 GB of video memory.

For even higher precision you can run models at the Q8_0 quant. StableLM 2 1.6B, Sana 1.6B, Zonos 0.1 1.6B, and Dia 1.6B all use 2 GB of video memory at Q8_0. Whisper Large v3 1.55B also fits at Q8_0 using 2 GB. Other Q8_0 options include ControlNet 1.5B, Hunyuan-DiT 1.5B, Stable Video Diffusion 1.5B, Whisper Large v2 1.5B, AudioGen 1.5B, and AudioLDM 2 1.5B which all use 1.9 GB of video memory.

When a model exceeds 2 GB you must use CPU offload. This process splits the model between your video memory and your system RAM. Offload allows you to run larger models but it slows down generation speeds significantly. For example SmolLM3 3B, Replit Code v1.5 3B, Kandinsky 3.1 3B, Voxtral Mini 3B, Orpheus TTS 3B, and Higgs Audio v2 3B need 2.2 GB of video memory at Q4_K_M and 4.2 GB of system RAM.

Larger offload cases include MusicGen 3.3B which needs 2.4 GB of video memory at Q4_K_M and 4.4 GB of system RAM. Stable Diffusion XL 3.417B needs 4.1 GB of video memory at FP8 and 6.1 GB of system RAM. SDXL Turbo 3.5B and SDXL Lightning 3.5B both need 2.6 GB of video memory at Q4_K_M and 4.6 GB of system RAM. Note that a 4k context window increases memory usage during generation so you must monitor your remaining video memory closely.