Best local AI models for Intel Arc A750

The largest models that fit fully

The 30 largest of the 123 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 Intel Arc A750 graphics card features 8 GB of GDDR6 video memory. This onboard memory size dictates which artificial intelligence models can run directly on your hardware. To run a model entirely on the graphics card, the model files and the active processing data must fit within this 8 GB limit. If a model exceeds this capacity, it cannot run solely on the graphics processor.

The quantization column indicates the compression level applied to each model. Quantization reduces the size of a model so it uses less memory. For example, the Q4_K_M quant represents a four bit quantization level, while Q5_K_M and Q6_K represent five bit and six bit levels. Higher quantization levels like Q6_K preserve more original model quality but require more memory. Lower levels like Q4_K_M allow larger models to fit into the 8 GB limit.

Several highly capable models fit entirely within the local memory of the Intel Arc A750. The largest fully fitting model is Mochi 1, a 10B model using the Q4_K_M quant which consumes 7.3 GB of memory. Other options include Gemma 2 9B at Q4_K_M using 8 GB, and Nemotron Nano 4B / 9B, GLM-4 9B / GLM-4.5-Air, Yi-Coder 1.5B / 9B, GLM-4-9B-Chat / CodeGeeX4, and GLM-4V-9B / GLM-4.1V-Thinking, which all use 7.7 GB at Q5_K_M. Chroma is an 8.9B model using 7.6 GB at Q5_K_M. Llama 3.1 8B fits at Q5_K_M using 7.4 GB.

You can also run 8B models at the higher quality Q6_K quantization level, which uses 7.9 GB of memory. These models include Granite 3.3 2B / 8B, Ministral 3B / 8B, InternLM 3 8B, OpenCoder 1.5B / 8B, Seed-Coder 8B, MiniCPM-V 2.6 / MiniCPM-o 2.6, Idefics 3 8B, Fuyu-8B, Emu3, and Stable Diffusion 3.5 Large / Turbo. EXAONE 3.5 2.4B / 7.8B uses 7.7 GB at Q6_K. Several 7B models fit easily at Q6_K using 6.9 GB, including Qwen2.5 0.5B / 1.5B / 3B / 7B, OLMo 2 1B / 7B, Falcon 3 1B / 3B / 7B, Command R7B, OpenHermes 2.5, Zephyr 7B Beta, OpenChat 3.5, Starling LM 7B, and Codestral Mamba 7B. Mistral 7B uses 7.4 GB at Q6_K.

When a model is too large for the 8 GB video memory, you must offload part of the workload to your system RAM. This offload process allows you to run larger models but reduces processing speed because system RAM is slower than GDDR6 memory. For offloading, we assume a system with 32 GB of system RAM. Under this setup, Open-Sora 2.0 is an 11B model needing 8.1 GB at Q4_K_M and 10.1 GB of system RAM. FLUX.1 dev is a 12B model needing 14.4 GB at FP8 / optimized and 16.4 GB of system RAM.

Other offload options include Gemma 3 12B, Gemma 4 12B, Mistral NeMo 12B, Pixtral 12B, FLUX.1 schnell, FLUX.1 Kontext dev, and FLUX.1 Krea dev. These 12B models all need 8.8 GB at Q4_K_M and 10.8 GB of system RAM. Vicuna 13B is a 13B model that needs 9.5 GB at Q4_K_M and 11.5 GB of system RAM. Note that all memory calculations are based on a standard 4k context window. If you increase the context window to process longer texts, the model will require significantly more memory and may no longer fit.