readygpu

Intel Arc Pro B70 vs NVIDIA GeForce RTX 5060 Ti 16GB

For running models locally, priced in Germany. At 64k context the first runs 58 models fully in memory, the second 31.

In short

The Intel Arc Pro B70 has more memory for models: 32 GB against 16 GB. At 64k context the Intel Arc Pro B70 runs 58 of the models we track fully in memory and the NVIDIA GeForce RTX 5060 Ti 16GB 31; only the Intel Arc Pro B70 holds Qwen 3.8 27B. On the models both run fully in memory, the Intel Arc Pro B70 is about 1.1× as fast as the NVIDIA GeForce RTX 5060 Ti 16GB (estimated, median across models).

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

Intel Arc Pro B70

Not priced here

Add to build
NVIDIA GeForce RTX 5060 Ti 16GB

720 €

eBay

Add to build

Specifications

SpecIntel Arc Pro B70NVIDIA GeForce RTX 5060 Ti 16GB
GPUArc Pro B70GeForce RTX 5060 Ti 16GB
Memory32 GB GDDR616 GB GDDR7
Memory bus256-bit128-bit
Memory bandwidth608 GB/s448 GB/s
FP16 compute45.8 TFLOPS47 TFLOPS
SoftwareSYCL / oneAPICUDA
Board power230 W180 W
PCIe generationGen 5Gen 5
PCIe lanesx16x8

What each runs

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

ModelIntel Arc Pro B70NVIDIA GeForce RTX 5060 Ti 16GB
Qwen 3.8 27B11tok/sQ6_Koffload
Qwen 3.6 27B11tok/sQ6_Koffload
Gemma 4 31B10tok/sQ6_Koffload
Qwen 3.6 35B-A3B43tok/sUD-Q6_Koffload
Qwen 3.5 27B11tok/sQ6_Koffload
Gemma 4 26B-A4B29tok/sQ8_0offload
Qwen 3.5 35B-A3B44tok/sQ6_Koffload
GLM 4.7 Flash22tok/sQ6_Koffload
Devstral Small 2 24B10tok/sQ6_Koffload
Nemotron 3.5 Lightning 30B-A3B50tok/sUD-Q5_K_Moffload
Kimi Linear 48B-A3B70tok/sQ4_K_Moffload
gpt-oss 20B41tok/sMXFP441tok/sMXFP4
Qwen3 Coder 30B-A3B19tok/sQ6_Koffload
Gemma 4 12B22tok/sQ8_020tok/sQ8_0
Qwen 3.5 9B26tok/sQ8_024tok/sQ8_0
Gemma 3 27B11tok/sQ6_Koffload
Mistral Small 3.2 24B10tok/sQ6_Koffload
Ministral 3 14B12tok/sQ8_0offload
Qwen3 14B12tok/sQ8_0offload
DeepSeek R1 0528 Qwen3 8B17tok/sQ8_019tok/sQ4_K_M
Qwen3 8B17tok/sQ8_019tok/sQ4_K_M
Phi-4 14B11tok/sQ8_0offload
Qwen 3.5 4B46tok/sQ8_042tok/sQ8_0
Ministral 3 8B17tok/sQ8_018tok/sQ5_K_M
Granite 4.2 8B15tok/sQ8_0offload
Gemma 4 E4B30tok/sQ8_028tok/sQ8_0
LFM2.5 8B-A1B39tok/sQ8_039tok/sQ8_0
Llama 3.1 8B20tok/sQ6_K18tok/sQ6_K
Qwen2.5 0.5B228tok/sQ8_0208tok/sQ8_0
Qwen3.5 0.8B186tok/sQ8_0170tok/sQ8_0
Qwen3.5 2B135tok/sQ5_K_M123tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B81tok/sQ8_074tok/sQ8_0
MiniCPM5 2B55tok/sQ8_051tok/sQ8_0
Spark X2.5 4B55tok/smlx-4bit50tok/smlx-4bit
Mellum2.1 12B A2.5B Thinking46tok/sQ8_046tok/sQ8_0
FrogNano 4B 260945tok/sQ8_041tok/sQ8_0
LFM2.5 2.6B44tok/sQ8_040tok/sQ8_0
LFM2.5 VL 3B44tok/sQ8_040tok/sQ8_0
granite vision 4.1 4b34tok/sQ8_031tok/sQ8_0
granite 4.2 3b33tok/sQ8_030tok/sQ8_0
Kimi VL A3B Thinking 250633tok/sQ8_042tok/sQ4_K_M
Kimi VL A3B32tok/smlx-8bit40tok/smlx-5bit
Ministral 3 3B Instruct 251229tok/sQ8_026tok/sQ8_0
GLM 4 9B 041428tok/smlx-5bit25tok/smlx-5bit
DeepSeek R1 Distill Qwen 7B26tok/sQ8_023tok/sQ8_0
GLM 4.6V Flash24tok/sQ8_022tok/sQ8_0
LFM2.5 24B A2B22tok/sQ8_0offload
GLM 4.1V 9B Thinking21tok/smlx-8bit19tok/smlx-8bit
DeepSeek R1 Distill Llama 8B18tok/sQ8_018tok/sQ6_K
Ministral 3 8B Reasoning 251217tok/sQ8_018tok/sQ5_K_M
Muse Glimmer 30B13tok/sQ6_Koffload
Ministral 3 14B Reasoning 251212tok/sQ8_0offload
Qwen3.8 27B OBLITERATED12tok/smlx-4bitoffload
Devstral Small 250711tok/sQ5_K_Moffload
Magistral Small 250711tok/sQ5_K_Moffload
Magistral Small 250911tok/sQ5_K_Moffload
DeepSeek R1 Distill Qwen 14B11tok/sQ8_0offload
Magistral Small 250610tok/sQ6_Koffload
Qwen 3.5 122B-A10Boffloadwon't run
Qwen3 Coder Nextoffloadwon't run
Nemotron 3 Super 120B-A12Boffloadwon't run
Mistral Small 4 119Boffloadoffload
GLM 4.5 Airoffloadwon't run
Qwen3 Next 80B-A3Boffloadwon't run
Qwen3 32Boffloadoffload
DeepSeek R1 Distill Qwen 32Boffloadoffload
granite 4.2 30boffloadoffload
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
gpt-oss 120Bwon't runwon't run
Llama 3.3 70Bwon't runwon't run
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Distill Llama 70Bwon't runwon't run
DeepSeek R1 Zerowon't runwon't run
DeepSeek V3won'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 4.5won'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 5won'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 B70 or the NVIDIA GeForce RTX 5060 Ti 16GB?
Choose the Intel Arc Pro B70 to run bigger models. Choose the Intel Arc Pro B70 for speed.
Which can run bigger models, the Intel Arc Pro B70 or the NVIDIA GeForce RTX 5060 Ti 16GB?
At 64k context the Intel Arc Pro B70 runs 58 of the models we track fully in memory and the NVIDIA GeForce RTX 5060 Ti 16GB 31; only the Intel Arc Pro B70 holds Qwen 3.8 27B.
Which is faster for local AI, the Intel Arc Pro B70 or the NVIDIA GeForce RTX 5060 Ti 16GB?
On the models both run fully in memory, the Intel Arc Pro B70 is about 1.1× as fast as the NVIDIA GeForce RTX 5060 Ti 16GB (estimated, median across models).

More comparisons

More hardware for local AI