readygpu

Intel Arc Pro B70 vs NVIDIA GeForce RTX 4070

For running models locally, priced in the US. At 8k context the first runs 62 models fully in memory, the second 35.

In short

The Intel Arc Pro B70 has more memory for models: 32 GB against 12 GB. At 8k context the Intel Arc Pro B70 runs 62 of the models we track fully in memory and the NVIDIA GeForce RTX 4070 35; only the Intel Arc Pro B70 holds Qwen 3.8 27B. On the models both run fully in memory, the Intel Arc Pro B70 and the NVIDIA GeForce RTX 4070 are about as fast as each other (estimated). In the US the Intel Arc Pro B70 costs $1,840 at Amazon and the NVIDIA GeForce RTX 4070 $570 at eBay: the NVIDIA GeForce RTX 4070 is $1,270 cheaper.

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

Intel Arc Pro B70

$1,840

Amazon

Add to build
NVIDIA GeForce RTX 4070

$570

eBay

Add to build

Specifications

SpecIntel Arc Pro B70NVIDIA GeForce RTX 4070
GPUArc Pro B70GeForce RTX 4070
Memory32 GB GDDR612 GB GDDR6X
Memory bus256-bit192-bit
Memory bandwidth608 GB/s504.2 GB/s
FP16 compute45.8 TFLOPS29.146 TFLOPS
SoftwareSYCL / oneAPICUDA
Board power230 W200 W
PCIe generationGen 5Gen 4
PCIe lanesx16x16

What each runs

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

ModelIntel Arc Pro B70NVIDIA GeForce RTX 4070
Qwen 3.8 27B10tok/sQ8_0offload
Qwen 3.6 27B10tok/sQ8_0offload
Gemma 4 31B11tok/sQ6_Koffload
Qwen 3.6 35B-A3B59tok/sQ6_Koffload
Qwen 3.5 27B10tok/sQ8_0offload
Gemma 4 26B-A4B37tok/sQ8_0offload
Qwen 3.5 35B-A3B63tok/sQ6_Koffload
GLM 4.7 Flash29tok/sQ8_0offload
Devstral Small 2 24B12tok/sQ8_0offload
Nemotron 3.5 Lightning 30B-A3B56tok/sUD-Q5_K_Moffload
Kimi Linear 48B-A3B86tok/sQ4_K_Moffload
gpt-oss 20B64tok/sMXFP4offload
Qwen3 Coder 30B-A3B51tok/sQ6_Koffload
Qwen3 32B11tok/sQ6_Koffload
Gemma 4 12B23tok/sQ8_031tok/sQ6_K
Qwen 3.5 9B31tok/sQ8_032tok/sQ8_0
Mistral Small 3.2 24B12tok/sQ8_0offload
DeepSeek R1 Distill Qwen 32B11tok/sQ6_Koffload
Gemma 3 27B10tok/sQ8_0offload
Ministral 3 14B19tok/sQ8_029tok/sQ5_K_M
Qwen3 14B18tok/sQ8_030tok/sQ4_K_M
DeepSeek R1 0528 Qwen3 8B31tok/sQ8_032tok/sQ8_0
Qwen3 8B31tok/sQ8_032tok/sQ8_0
Phi-4 14B18tok/sQ8_030tok/sQ4_K_M
Qwen 3.5 4B63tok/sQ8_065tok/sQ8_0
Ministral 3 8B30tok/sQ8_031tok/sQ8_0
Granite 4.2 8B28tok/sQ8_029tok/sQ8_0
Gemma 4 E4B36tok/sQ8_037tok/sQ8_0
LFM2.5 8B-A1B114tok/sQ8_0128tok/sQ8_0
Llama 3.1 8B40tok/sQ6_K41tok/sQ6_K
Qwen2.5 0.5B481tok/sQ8_0495tok/sQ8_0
Qwen3.5 0.8B326tok/sQ8_0336tok/sQ8_0
Qwen3.5 2B196tok/sQ5_K_M201tok/sQ5_K_M
LFM3 1.7B155tok/sQ8_0160tok/sQ8_0
DeepSeek R1 Distill Qwen 1.5B143tok/sQ8_0147tok/sQ8_0
MiniCPM5 2B100tok/sQ8_0103tok/sQ8_0
LFM2.5 2.6B90tok/sQ8_093tok/sQ8_0
LFM2.5 VL 3B90tok/sQ8_093tok/sQ8_0
Spark X2.5 4B89tok/smlx-4bit92tok/smlx-4bit
granite vision 4.1 4b71tok/sQ8_073tok/sQ8_0
Ministral 3 3B Instruct 251267tok/sQ8_069tok/sQ8_0
granite 4.2 3b67tok/sQ8_069tok/sQ8_0
FrogNano 4B 260962tok/sQ8_063tok/sQ8_0
LFM2.5 24B A2B60tok/sQ8_0offload
Mellum2.1 12B A2.5B Thinking59tok/sQ8_085tok/sQ5_K_M
Kimi VL A3B Thinking 250650tok/sQ8_086tok/sQ4_K_M
Kimi VL A3B48tok/smlx-8bitoffload
DeepSeek R1 Distill Qwen 7B36tok/sQ8_037tok/sQ8_0
GLM 4 9B 041435tok/smlx-5bit36tok/smlx-5bit
DeepSeek R1 Distill Llama 8B32tok/sQ8_033tok/sQ8_0
Ministral 3 8B Reasoning 251230tok/sQ8_031tok/sQ8_0
GLM 4.6V Flash29tok/sQ8_030tok/sQ8_0
GLM 4.1V 9B Thinking25tok/smlx-8bitoffload
Ministral 3 14B Reasoning 251219tok/sQ8_029tok/sQ5_K_M
DeepSeek R1 Distill Qwen 14B18tok/sQ8_030tok/sQ4_K_M
Qwen3.8 27B OBLITERATED14tok/smlx-4bitoffload
Devstral Small 250712tok/sQ8_0offload
Magistral Small 250612tok/sQ8_0offload
Magistral Small 250712tok/sQ8_0offload
Magistral Small 250912tok/sQ8_0offload
granite 4.2 30b11tok/sQ6_Koffload
Muse Glimmer 30B9tok/sQ8_0offload
Qwen 3.5 122B-A10Boffloadwon't run
Qwen3 Coder Nextoffloadwon't run
Nemotron 3 Super 120B-A12Boffloadwon't run
Mistral Small 4 119Boffloadwon't run
GLM 4.5 Airoffloadwon't run
Qwen3 Next 80B-A3Boffloadwon't run
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
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 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 4070?
Choose the Intel Arc Pro B70 to run bigger models. Choose the NVIDIA GeForce RTX 4070 to spend less.
Which can run bigger models, the Intel Arc Pro B70 or the NVIDIA GeForce RTX 4070?
At 8k context the Intel Arc Pro B70 runs 62 of the models we track fully in memory and the NVIDIA GeForce RTX 4070 35; 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 4070?
On the models both run fully in memory, the Intel Arc Pro B70 and the NVIDIA GeForce RTX 4070 are about as fast as each other (estimated).
What is the price difference between the Intel Arc Pro B70 and the NVIDIA GeForce RTX 4070?
In the US the Intel Arc Pro B70 costs $1,840 at Amazon and the NVIDIA GeForce RTX 4070 $570 at eBay: the NVIDIA GeForce RTX 4070 is $1,270 cheaper.

More comparisons

More hardware for local AI