readygpu

Intel Arc Pro B60 vs NVIDIA GeForce RTX 3080 10GB

For running models locally, priced in Canada. At 8k context the first runs 56 models fully in memory, the second 26.

In short

The Intel Arc Pro B60 has more memory for models: 24 GB against 10 GB. At 8k context the Intel Arc Pro B60 runs 56 of the models we track fully in memory and the NVIDIA GeForce RTX 3080 10GB 26; only the Intel Arc Pro B60 holds Qwen 3.8 27B. On the models both run fully in memory, the NVIDIA GeForce RTX 3080 10GB is about 2.1× as fast as the Intel Arc Pro B60 (estimated, median across models). In Canada the Intel Arc Pro B60 costs $1,057 at Amazon and the NVIDIA GeForce RTX 3080 10GB $850 at eBay: the NVIDIA GeForce RTX 3080 10GB is $207 cheaper.

Updated . Speeds are estimates; how they are worked out.

Intel Arc Pro B60

$1,057

Amazon

Add to build
NVIDIA GeForce RTX 3080 10GB

$850

eBay

Add to build

Specifications

SpecIntel Arc Pro B60NVIDIA GeForce RTX 3080 10GB
GPUArc Pro B60GeForce RTX 3080 10GB
Memory24 GB GDDR610 GB GDDR6X
Memory bus192-bit320-bit
Memory bandwidth456 GB/s760 GB/s
FP16 compute98.5 TFLOPS59.5 TFLOPS
SoftwareSYCL / oneAPICUDA
Board power200 W320 W
PCIe generationGen 5Gen 4
PCIe lanesx8x16

What each runs

Estimated decode speed with the most precise quant that fits each one est.

ModelIntel Arc Pro B60NVIDIA GeForce RTX 3080 10GB
Qwen 3.8 27B10tok/sUD-Q6_Koffload
Qwen 3.6 27B10tok/sQ6_Koffload
Gemma 4 31B10tok/sQ5_K_Moffload
Qwen 3.6 35B-A3B60tok/sQ4_K_Moffload
Qwen 3.5 27B10tok/sQ6_Koffload
Gemma 4 26B-A4B32tok/sUD-Q6_Koffload
Qwen 3.5 35B-A3B61tok/sQ4_K_Moffload
GLM 4.7 Flash31tok/sQ5_K_Moffload
Devstral Small 2 24B11tok/sQ6_Koffload
Nemotron 3.5 Lightning 30B-A3B54tok/sMXFP4offload
gpt-oss 20B48tok/sMXFP4offload
Qwen3 Coder 30B-A3B43tok/sQ5_K_Moffload
Qwen3 32B10tok/sQ4_K_Moffload
Gemma 4 12B17tok/sQ8_053tok/sQ5_K_M
Qwen 3.5 9B23tok/sQ8_061tok/sQ6_K
Mistral Small 3.2 24B11tok/sQ6_Koffload
DeepSeek R1 Distill Qwen 32B10tok/sQ4_K_Moffload
Gemma 3 27B10tok/sQ6_Koffload
Ministral 3 14B15tok/sQ8_0offload
Qwen3 14B13tok/sQ8_0offload
DeepSeek R1 0528 Qwen3 8B23tok/sQ8_059tok/sQ6_K
Qwen3 8B23tok/sQ8_059tok/sQ6_K
Phi-4 14B13tok/sQ8_0offload
Qwen 3.5 4B48tok/sQ8_098tok/sQ8_0
Ministral 3 8B22tok/sQ8_058tok/sQ6_K
Granite 4.2 8B21tok/sQ8_053tok/sQ6_K
Gemma 4 E4B27tok/sQ8_056tok/sQ8_0
LFM2.5 8B-A1B86tok/sQ8_0228tok/sUD-Q6_K
Llama 3.1 8B30tok/sQ6_K61tok/sQ6_K
Qwen2.5 0.5B361tok/sQ8_0746tok/sQ8_0
Qwen3.5 0.8B245tok/sQ8_0506tok/sQ8_0
Qwen3.5 2B147tok/sQ5_K_M303tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B107tok/sQ8_0221tok/sQ8_0
MiniCPM5 2B75tok/sQ8_0155tok/sQ8_0
LFM2.5 2.6B68tok/sQ8_0140tok/sQ8_0
LFM2.5 VL 3B68tok/sQ8_0140tok/sQ8_0
Spark X2.5 4B67tok/smlx-4bit139tok/smlx-4bit
granite vision 4.1 4b53tok/sQ8_0110tok/sQ8_0
LFM2.5 24B A2B52tok/sQ6_Koffload
granite 4.2 3b50tok/sQ8_0103tok/sQ8_0
Mellum2.1 12B A2.5B Thinking45tok/sQ8_0143tok/sQ4_K_M
Kimi VL A3B Thinking 250637tok/sQ8_0offload
Kimi VL A3B36tok/smlx-8bitoffload
DeepSeek R1 Distill Qwen 7B27tok/sQ8_055tok/sQ8_0
GLM 4 9B 041427tok/smlx-5bit55tok/smlx-5bit
DeepSeek R1 Distill Llama 8B24tok/sQ8_061tok/sQ6_K
Ministral 3 8B Reasoning 251222tok/sQ8_065tok/sQ5_K_M
GLM 4.6V Flash22tok/sQ8_073tok/sQ4_K_M
GLM 4.1V 9B Thinking19tok/smlx-8bitoffload
DeepSeek R1 Distill Qwen 14B13tok/sQ8_0offload
Devstral Small 250713tok/sQ5_K_Moffload
Magistral Small 250713tok/sQ5_K_Moffload
Magistral Small 250913tok/sQ5_K_Moffload
Magistral Small 250611tok/sQ6_Koffload
Qwen3.8 27B OBLITERATED11tok/smlx-4bitoffload
granite 4.2 30b10tok/sQ5_K_Moffload
Qwen 3.5 122B-A10Boffloadwon't run
Qwen3 Coder Nextoffloadwon't run
Mistral Small 4 119Boffloadwon't run
GLM 4.5 Airoffloadwon't run
Qwen3 Next 80B-A3Boffloadwon't run
Kimi Linear 48B-A3Boffloadoffload
Llama 3.3 70Boffloadwon't run
DeepSeek R1 Distill Llama 70Boffloadwon't run
DeepSeek V4 Flashwon't runwon't run
GLM 5.3 Flashwon't runwon't run
Qwen 3.8 Flash Nextwon't runwon't run
MiniMax M2.7won't runwon't run
Nemotron 3 Super 120B-A12Bwon't runwon't run
gpt-oss 120Bwon't runwon't run
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Zerowon't runwon't run
DeepSeek V3 0324won't runwon't run
DeepSeek V3.1won't runwon't run
DeepSeek V3.1 Terminuswon't runwon't run
DeepSeek V3.2won't runwon't run
DeepSeek V4 Flash 0731won't runwon't run
DeepSeek V4 Flash Vision Expwon't runwon't run
Devstral 2 123B Instruct 2512won't runwon't run
GLM 4.5Vwon't runwon't run
GLM 4.6won't runwon't run
GLM 4.6Vwon't runwon't run
GLM 4.7won't runwon't run
GLM 5.1won't runwon't run
GLM 5.2won't runwon't run
GLM 5.3won't runwon't run
Kimi Dev 72Bwon't runwon't run
Kimi K2won't runwon't run
Kimi K2 Instruct 0905won't runwon't run
Kimi K2 Thinkingwon't runwon't run
Kimi K2.5won't runwon't run
MiniMax M2won't runwon't run
MiniMax M2.1won't runwon't run
MiniMax M2.5won't runwon't run
MiniMax M3won't runwon't run
Mistral Medium 3.5 128Bwon't runwon't run

Questions

Which is better for local AI, the Intel Arc Pro B60 or the NVIDIA GeForce RTX 3080 10GB?
Choose the Intel Arc Pro B60 to run bigger models. Choose the NVIDIA GeForce RTX 3080 10GB for speed. Choose the NVIDIA GeForce RTX 3080 10GB to spend less.
Which can run bigger models, the Intel Arc Pro B60 or the NVIDIA GeForce RTX 3080 10GB?
At 8k context the Intel Arc Pro B60 runs 56 of the models we track fully in memory and the NVIDIA GeForce RTX 3080 10GB 26; only the Intel Arc Pro B60 holds Qwen 3.8 27B.
Which is faster for local AI, the Intel Arc Pro B60 or the NVIDIA GeForce RTX 3080 10GB?
On the models both run fully in memory, the NVIDIA GeForce RTX 3080 10GB is about 2.1× as fast as the Intel Arc Pro B60 (estimated, median across models).
What is the price difference between the Intel Arc Pro B60 and the NVIDIA GeForce RTX 3080 10GB?
In Canada the Intel Arc Pro B60 costs $1,057 at Amazon and the NVIDIA GeForce RTX 3080 10GB $850 at eBay: the NVIDIA GeForce RTX 3080 10GB is $207 cheaper.

More comparisons