Best local AI models for Intel Arc Pro B60

The largest models that fit fully

The 30 largest of the 173 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 Pro B60 workstation graphics card features 24 GB of GDDR6 dedicated video memory. This memory size determines the maximum size of the artificial intelligence models you can run locally. To fit a model entirely within the onboard graphics memory, the combined weight of the model parameters and the active context window must not exceed this 24 GB limit.

The quantization column indicates the optimal compression format for each model. Quantization reduces the precision of model weights to save space. For example, the Qwen3 32B, Qwen3.5 dense variants, Aya Expanse 32B, Granite 4.0 Small or Tiny, and Qwen2.5-Coder 32B models all fit within 23.4 GB of video memory using the Q4_K_M quantization. Other models like Gemma 3 27B, Gemma 3 vision, Wan 2.2 or 2.5, Gemma 4 26B-A4B, and Aria use the Q5_K_M quantization to fit within their respective memory footprints.

Higher precision quantizations are available for slightly smaller models. Mistral Small 3.2, Magistral Small, and Devstral Small 1.1 fit within 23.6 GB using the Q6_K quantization. Solar Pro and Codestral 22B also use Q6_K to fit within 21.6 GB. For maximum precision, models like HunyuanImage 2.1 or 3.0, Ling-Coder-Lite, DeepSeek-Coder-V2 16B, Kimi-VL A3B, Apriel-1.5-15B-Thinker, StarCoder2 15B, and Qwen2.5 14B utilize the Q8_0 quantization format.

When a model exceeds the 24 GB video memory of the Intel Arc Pro B60, you must offload parts of the model to your system RAM. This offloading process allows you to run larger models but reduces processing speed because system RAM is slower than graphics memory. For these offload cases, we assume your computer has 32 GB of system RAM available to handle the overflow.

Examples of offloaded models include OTel 2.0 LLM 31B IT, which needs 27.5 GB of memory at Q4_K_M quantization and requires 29.5 GB of system RAM. DeepSeek-Coder 33B and WizardCoder 33B need 24.2 GB of memory at Q4_K_M and require 26.2 GB of system RAM. Similarly, Yi 1.5 34B, Granite Code 34B, LLaVA 1.5 or 1.6 34B, and Ovis 2 need 24.9 GB at Q4_K_M and require 26.9 GB of system RAM. Qwen3.6-35B-A3B and Command R 35B need 25.6 GB at Q4_K_M and require 27.6 GB of system RAM, while Seed-OSS 36B needs 26.4 GB at Q4_K_M and requires 28.4 GB of system RAM.

All memory calculations on this page assume a standard 4k context window. If you increase the context window to process longer documents or larger chat histories, the model will require more memory. This extra memory requirement may force you to use a lower quantization level or offload more layers to your system RAM.