Intel's fastest current gaming card, a $249 card for 1080p and 1440p with 12 GB of memory: 56.5 fps in Cyberpunk 2077 at 1440p Ultra without upscaling in KitGuru's testing. It offers more memory than rivals at its price and runs ray tracing better than AMD's older budget cards, but TechSpot found Arc loses more performance than rivals when paired with a slower processor.
Against the line-up
Cyberpunk 2077
Cyberpunk 2077
Cyberpunk 2077
Black Myth: Wukong
Local LLM (llama.cpp / Ollama)
Local LLM (llama.cpp / Ollama)
Blender Benchmark
The whole Graphics cards line-up side by side, with prices
Every benchmark
Quoted from the publications that ran them; we don't test hardware ourselves. Figures from different publications were run on different systems and scenes, so compare within a row's source.
Cyberpunk 2077
| Setting | Result | Source |
|---|---|---|
| 1080p Ultra |
84.1fps | KitGuru, Jan 16, 2025 Ryzen 7 9800X3D, 64GB DDR5-6000 CL30, MSI MPG X870E Carbon WiFi (KitGuru GPU test system) Card tested: Intel Arc B580 Limited Edition 12GB. Ultra preset, FSR/DLSS/XeSS disabled, DX12. Average fps. |
| 1440p High |
63fps | TechSpot (Hardware Unboxed), Jun 4, 2025 Ryzen 7 9800X3D, 32GB DDR5-6000 CL30, MSI MPG X870E Carbon WiFi (TechSpot) Phantom Liberty, native |
| 1440p RT Ultra |
23.1fps | Tom's Hardware, Dec 13, 2024 Ryzen 7 9800X3D, 32GB DDR5-6000, ASRock X670E Taichi (Tom's Hardware 2025 GPU testbed) Ray Tracing Ultra preset, native |
| 1440p Ultra |
56.5fps | KitGuru, Jan 16, 2025 Ryzen 7 9800X3D, 64GB DDR5-6000 CL30, MSI MPG X870E Carbon WiFi (KitGuru GPU test system) Card tested: Intel Arc B580 Limited Edition 12GB. Ultra preset, FSR/DLSS/XeSS disabled, DX12. Average fps. |
| 1440p RT Ultra, Quality upscaling |
41fps | TechSpot (Hardware Unboxed), Jun 4, 2025 Ryzen 7 9800X3D, 32GB DDR5-6000 CL30, MSI MPG X870E Carbon WiFi (TechSpot) Phantom Liberty; Ray Tracing: Ultra with Quality-mode upscaling |
| 4K Ultra |
27.3fps | KitGuru, Jan 16, 2025 Ryzen 7 9800X3D, 64GB DDR5-6000 CL30, MSI MPG X870E Carbon WiFi (KitGuru GPU test system) Card tested: Intel Arc B580 Limited Edition 12GB. Ultra preset, FSR/DLSS/XeSS disabled, DX12. Average fps. |
Alan Wake 2
| Setting | Result | Source |
|---|---|---|
| 1440p High, RT High, Quality upscaling |
16fps | TechSpot (Hardware Unboxed), Jun 4, 2025 Ryzen 7 9800X3D, 32GB DDR5-6000 CL30, MSI MPG X870E Carbon WiFi (TechSpot) High preset, ray tracing High, Quality-mode upscaling |
Black Myth: Wukong
| Setting | Result | Source |
|---|---|---|
| 1440p Very High |
21.6fps | Tom's Hardware, Dec 13, 2024 Ryzen 7 9800X3D, 32GB DDR5-6000, ASRock X670E Taichi (Tom's Hardware 2025 GPU testbed) no RT, DX12 |
Counter-Strike 2
| Setting | Result | Source |
|---|---|---|
| 1440p Medium |
263fps | TechSpot (Hardware Unboxed), Jun 4, 2025 Ryzen 7 9800X3D, 32GB DDR5-6000 CL30, MSI MPG X870E Carbon WiFi (TechSpot) |
Baldur's Gate 3
| Setting | Result | Source |
|---|---|---|
| 1440p Ultra |
71fps | Tom's Hardware, Dec 13, 2024 Ryzen 7 9800X3D, 32GB DDR5-6000, ASRock X670E Taichi (Tom's Hardware 2025 GPU testbed) DX11 |
Local LLM (llama.cpp / Ollama)
| Setting | Result | Source |
|---|---|---|
| Llama 2 7B (MLPerf Client 0.5), Generation |
80.5tok/s | Tom's Hardware, Dec 13, 2024 Tom's Hardware GPU test PC (the review used a Ryzen 7 9800X3D and a Core i9-13900K system) MLPerf Client 0.5, geomean of its four tests, tokens/s after the first token |
| pp512 Llama 2 7B Q4_0, Prompt processing |
621tok/s | llama.cpp community benchmarks (GitHub), Oct 19, 2026 llama-bench on Vulkan, flash attention off; scoreboard entry build 7f76692 |
| tg128 Llama 2 7B Q4_0, Generation |
70.1tok/s | llama.cpp community benchmarks (GitHub), Oct 19, 2026 llama-bench on Vulkan, flash attention off; scoreboard entry build 7f76692 |
UL Procyon AI Text Generation
| Setting | Result | Source |
|---|---|---|
| Llama 3.1 |
1,535points | KitGuru, Jun 4, 2025 Ryzen 7 9800X3D, 64GB DDR5-6000 CL30, MSI MPG X870E Carbon WiFi (KitGuru GPU test system) Card tested: Intel Arc B580 Limited Edition 12GB. UL Procyon AI Text Generation, ONNX Runtime with DirectML, Llama 3.1 overall score. |
Blender Benchmark
| Setting | Result | Source |
|---|---|---|
| Standard run |
2,373samples/min | Blender Open Data, Oct 19, 2026 Blender 4.5.0, ONEAPI, median of 124 submitted runs (monster, junkshop, classroom) |
| Blender 4.3 (oneAPI), geomean of 3 scenes |
583samples/min | Tom's Hardware, Dec 13, 2024 Tom's Hardware GPU test PC (the review used a Ryzen 7 9800X3D and a Core i9-13900K system) Monster, Junkshop, Classroom |
Sources
- Blender Open Data (results database)
- KitGuru (review)
- TechSpot (Hardware Unboxed) (review)
- Tom's Hardware (review)
- llama.cpp community benchmarks (GitHub) (results database)
As an Amazon Associate we earn from qualifying purchases, and we may earn from other shop links. Amazon's price is on Amazon; Best Buy's is read from Best Buy and shown with the time it was checked.