AMD Instinct MI300X vs Intel Arc Pro B60 Comparison

AMD
RADEON

AMD Instinct MI300X

CORE STATE Aqua Vanjaram
VRAM 192 GB
CLOCK SPEED 2100 MHz
TDP 750 W
BUS WIDTH 8192 bit
ARCHITECTURE CDNA 3.0
nm
PROCESS 5 nm
LAUNCH DATE 2023
VS
Intel
GPU

Arc Pro B60

CORE STATE BMG-G21
VRAM 24 GB
CLOCK SPEED 2400 MHz
TDP 200 W
BUS WIDTH 192 bit
ARCHITECTURE Xe2-HPG
nm
PROCESS 5 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

geekbench_opencl
317,994
N/A
3dmark_3dmark_steel_nomad_dx12
N/A
2,646
passmark_directx_10
N/A
61
passmark_directx_11
N/A
122
passmark_directx_12
N/A
76
passmark_directx_9
N/A
179
passmark_g2d
N/A
763
passmark_g3d
N/A
14,580
passmark_gpu_compute
N/A
7,029

Analysis: AMD Instinct MI300X vs Intel Arc Pro B60

Where Each One Wins

The benchmark data draws a stark line between these two accelerators. The AMD Instinct MI300X is a compute-first data center part that dominates in raw OpenCL throughput, while the Intel Arc Pro B60 is a workstation graphics card with a broader but much lower performance envelope.

The MI300X records a Geekbench OpenCL score of 317994, placing it in the 100th percentile of all GPUs in the database. Its nearest rivals show the context: it sits 5% behind the NVIDIA H200 NVL (334891) and 8% behind the NVIDIA B200 (345482), but it leads the NVIDIA L40S (295763) by 7.5% and the NVIDIA RTX 6000 Ada Generation (287237) by 10.7%. This is a compute accelerator engineered for scale, and the data confirms that positioning.

The Intel Arc Pro B60, by contrast, posts an average benchmark score of 3182 across a suite of DirectX and PassMark tests. Its percentile rank is 20, meaning the vast majority of GPUs in the database outperform it. Its closest rivals in the database are modest parts: the NVIDIA Quadro P1000 (3163, 0.6% behind), the NVIDIA GeForce GT 640 (3210, 0.9% ahead), and the NVIDIA GeForce 920M (3287, 3.2% ahead). Even the NVIDIA GeForce RTX 5080 SUPER (3075) is only 3.5% behind the Arc Pro B60 in this average. The Arc Pro B60 wins no head-to-head benchmark comparisons against the MI300X, and the MI300X wins none against it, because the two share no common test in the database.

The use-case split is therefore clear. The MI300X wins where massive compute throughput matters, specifically in OpenCL workloads that scale to hundreds of thousands of operations. The Arc Pro B60 wins where graphical API support and display output matter, but its raw compute scores are orders of magnitude lower. Data indicates the MI300X is a server-class compute engine, while the Arc Pro B60 is a workstation display adapter with secondary compute capabilities.

Architecture Differences

The two chips come from different architectural lineages. The AMD Instinct MI300X uses the CDNA 3.0 architecture, built on a 5 nm TSMC process with 153,000 million transistors on a 1017 mm² die. The transistor density is 150.4 million per mm². The Intel Arc Pro B60 uses the Xe2-HPG architecture, also on a 5 nm TSMC process, but with 19,600 million transistors on a 272 mm² die, yielding a density of 72.1 million per mm². The MI300X is a much larger and denser chip.

The compute resources differ massively. The MI300X has 19,456 shading units and 1,216 texture mapping units, with no ROPs and no dedicated ray tracing cores. The Arc Pro B60 has 2,560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores. The MI300X reports a pixel rate of 0 MPixel/s, while the Arc Pro B60 delivers 192.0 GPixel/s. Texture rates are 2,553.6 GTexel/s for the MI300X versus 384.0 GTexel/s for the Arc Pro B60. FP32 throughput is 81.72 TFLOPS for the AMD part versus 12.29 TFLOPS for the Intel part. The MI300X delivers FP16 at 81.72 TFLOPS (1:1 ratio), while the Arc Pro B60 delivers 24.58 TFLOPS (2:1 ratio).

Memory architecture is equally divergent. The MI300X uses 192 GB of HBM3 on an 8192-bit bus, with 5.32 TB/s of bandwidth and memory clocks of 1300 MHz (5.2 Gbps effective). The Arc Pro B60 uses 24 GB of GDDR6 on a 192-bit bus, with 456.0 GB/s of bandwidth and memory clocks of 2375 MHz (19 Gbps effective). The AMD part has over eleven times the memory capacity and over eleven times the bandwidth.

Clock behavior differs as well. The MI300X has a base clock of 1000 MHz and a boost clock of 2100 MHz. The Arc Pro B60 has a base clock of 2000 MHz and a boost clock of 2400 MHz. Despite lower clocks, the MI300X achieves far higher throughput due to the massive shader count and memory subsystem.

The MI300X is an OAM Module with no display outputs and no power connectors, requiring a suggested 1150 W PSU and rated at 750 W TDP. The Arc Pro B60 is a dual-slot card with 4x mini-DisplayPort 2.1 outputs, a single 8-pin power connector, a 200 W TDP, and a suggested 550 W PSU. The Intel part measures 167 mm by 69 mm by 40 mm. The AMD part has no listed dimensions.

API support is another separator. The MI300X reports N/A for DirectX, OpenGL, and Vulkan. The Arc Pro B60 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

The Verdict

The data supports a straightforward conclusion. The AMD Instinct MI300X is for compute workloads that demand enormous memory capacity, bandwidth, and raw FP32/FP16 throughput. Its 100th percentile OpenCL score and position among top NVIDIA data center parts confirm this. It has no display outputs and no graphics API support, so it is not a graphics card in any conventional sense.

The Intel Arc Pro B60 is for workstation graphics tasks. It provides display output through four mini-DisplayPort 2.1 connectors, supports modern graphics APIs including DirectX 12 Ultimate and Vulkan 1.4, and includes ray tracing hardware. Its compute scores are modest, with a 20th percentile rank and an average score of 3182. It is not competitive with the MI300X in raw compute, nor is it designed to be.

Users selecting between these two should base the choice on workload type. For large-scale compute, scientific simulation, or AI inference with massive memory footprints, the MI300X is the only option of the two. For workstation graphics, display output, and API-complete rendering, the Arc Pro B60 is the only option of the two. There is no overlap in their intended use cases.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The AMD Instinct MI300X scores 317994 in Geekbench OpenCL, placing it in the 100th percentile of all GPUs. The Intel Arc Pro B60 does not have an OpenCL benchmark record in the database.

Q: How does the MI300X compare to its nearest rivals?

A: The MI300X is 5% behind the NVIDIA H200 NVL (334891), 8% behind the NVIDIA B200 (345482), 7.5% ahead of the NVIDIA L40S (295763), and 10.7% ahead of the NVIDIA RTX 6000 Ada Generation (287237).

Q: What is the memory capacity difference?

A: The MI300X has 192 GB of HBM3 memory on an 8192-bit bus with 5.32 TB/s bandwidth. The Arc Pro B60 has 24 GB of GDDR6 memory on a 192-bit bus with 456.0 GB/s bandwidth.

Q: Does the Arc Pro B60 support ray tracing?

A: Yes, the Arc Pro B60 has 20 ray tracing cores. The MI300X has no dedicated ray tracing cores listed in the database.

Q: What display outputs does each card have?

A: The MI300X has no display outputs. The Arc Pro B60 has 4x mini-DisplayPort 2.1 outputs.

Q: What are the power requirements?

A: The MI300X is rated at 750 W TDP with a suggested 1150 W PSU and no power connectors (OAM Module). The Arc Pro B60 is rated at 200 W TDP with a suggested 550 W PSU and one 8-pin power connector.

Head-to-Head Benchmarks

The database contains no shared benchmark tests between the AMD Instinct MI300X and the Intel Arc Pro B60. The MI300X has a single benchmark record: Geekbench OpenCL at 317994. The Arc Pro B60 has eight records: 3DMark Steel Nomad DX12 at 2646, PassMark DirectX 10 at 61, PassMark DirectX 11 at 122, PassMark DirectX 12 at 76, PassMark DirectX 9 at 179, PassMark G2D at 763, PassMark G3D at 14580, and PassMark GPU Compute at 7029.

Despite the lack of direct comparison, the data allows indirect inference. The MI300X's OpenCL score of 317994 is an order of magnitude above the Arc Pro B60's strongest PassMark result (G3D at 14580) and its 3DMark score (2646). The MI300X's FP32 throughput of 81.72 TFLOPS is roughly 6.6 times the Arc Pro B60's 12.29 TFLOPS. Texture rate is 2,553.6 GTexel/s versus 384.0 GTexel/s, a factor of 6.65. Memory bandwidth is 5.32 TB/s versus 456.0 GB/s, a factor of 11.7.

The Arc Pro B60 counters in areas the MI300X does not address. Pixel rate is 192.0 GPixel/s versus 0 MPixel/s. The Arc Pro B60 has 20 ray tracing cores, while the MI300X has none. The Arc Pro B60 supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4; the MI300X lists N/A for all three APIs. The Arc Pro B60 provides 4x mini-DisplayPort 2.1 outputs; the MI300X provides none.

In average benchmark score, the MI300X records 317994, while the Arc Pro B60 records 3182. The percentile ranks are 100 versus 20. These aggregate metrics confirm that the MI300X is a dominant compute part, while the Arc Pro B60 sits in the lower quartile of all GPUs in the database.

Specification Differences

The two parts differ across nearly every specification field.

Process and die: Both use 5 nm TSMC, but the MI300X has 153,000 million transistors on a 1017 mm² die with 150.4M transistors per mm². The Arc Pro B60 has 19,600 million transistors on a 272 mm² die with 72.1M transistors per mm².

Clocks: The MI300X runs at 1000 MHz base and 2100 MHz boost, with memory at 1300 MHz (5.2 Gbps effective). The Arc Pro B60 runs at 2000 MHz base and 2400 MHz boost, with memory at 2375 MHz (19 Gbps effective).

Memory: The MI300X uses 192 GB of HBM3 on an 8192-bit bus with 5.32 TB/s bandwidth. The Arc Pro B60 uses 24 GB of GDDR6 on a 192-bit bus with 456.0 GB/s bandwidth.

Compute units: The MI300X has 19,456 shading units, 1,216 TMUs, 0 ROPs, and no ray tracing cores. The Arc Pro B60 has 2,560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores.

Rates: The MI300X has a pixel rate of 0 MPixel/s and a texture rate of 2,553.6 GTexel/s. The Arc Pro B60 has a pixel rate of 192.0 GPixel/s and a texture rate of 384.0 GTexel/s.

Compute throughput: The MI300X delivers 81.72 TFLOPS FP32 and 81.72 TFLOPS FP16 (1:1). The Arc Pro B60 delivers 12.29 TFLOPS FP32 and 24.58 TFLOPS FP16 (2:1).

Power and form factor: The MI300X is rated at 750 W TDP, uses an OAM Module form factor, has no power connectors, and suggests a 1150 W PSU. The Arc Pro B60 is rated at 200 W TDP, uses a dual-slot form factor, has one 8-pin power connector, and suggests a 550 W PSU.

Interfaces and outputs: The MI300X uses PCIe 5.0 x16 and has no display outputs. The Arc Pro B60 uses PCIe 5.0 x8 and has 4x mini-DisplayPort 2.1.

API support: The MI300X lists N/A for DirectX, OpenGL, and Vulkan. The Arc Pro B60 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Physical dimensions: The MI300X has no listed dimensions. The Arc Pro B60 measures 167 mm in length, 69 mm in height, and 40 mm in width.

Release and status: The MI300X was released on 2023-12-05 and has no listed production status or successor. The Arc Pro B60 was released on 2025-09-04, has an active production status, and no listed predecessor or successor. The Arc Pro B60 has a launch MSRP of 499 USD.

DETAILED SPECIFICATIONS

SPECIFICATION
Instinct MI300X
Pro B60
Core Specs
Shading Units
19,456
2,560 -86.8%
Shaders
19,456
2,560 -86.8%
TMUs
1,216
160 -86.8%
ROPs
0
80 +∞%
Compute Units
304
Execution Units
20
Clocks
Base Clock
1000 MHz
2000 MHz
Boost Clock
2100 MHz
2400 MHz
Memory Clock
1300 MHz 5.2 Gbps effective
2375 MHz 19 Gbps effective
Memory
Memory Size
192 GB
24 GB
VRAM (MB)
196,608
24,576 -87.5%
Memory Type
HBM3
GDDR6
Memory Bus
8192 bit
192 bit
Bandwidth
5.32 TB/s
456.0 GB/s
Cache
L1 Cache
16 KB (per CU)
L2 Cache
16 MB
10 MB
L3 Cache
256 MB
Performance
Pixel Rate
0 MPixel/s
192.0 GPixel/s
Texture Rate
2,553.6 GTexel/s
384.0 GTexel/s
FP32 (TFLOPS)
81.72 TFLOPS
12.29 TFLOPS
FP64 (TFLOPS)
40.86 TFLOPS (1:2)
3.072 TFLOPS (1:4)
FP16 (TFLOPS)
81.72 TFLOPS (1:1)
24.58 TFLOPS (2:1)
AI/RT
RT Cores
20
XMX Cores
160
Matrix Cores
1,216
Power
TDP
750 W
200 W
TDP (W)
750
200 -73.3%
Suggested PSU
1150 W
550 W
Power Connectors
None
1x 8-pin
Architecture
Architecture
CDNA 3.0
Xe2-HPG
GPU Name
Aqua Vanjaram
BMG-G21
Generation
Instinct (MIx)
Battlemage (Pro Series)
Process Size
5 nm
5 nm
Transistors
153,000 million
19,600 million
Die Size
1017 mm²
272 mm²
Foundry
TSMC
TSMC
Density
150.4M / mm²
72.1M / mm²
AMD MCM
MCM
2
API Support
DirectX
12 Ultimate (12_2)
OpenGL
4.6
Vulkan
1.4
OpenCL
3.0
3.0
Shader Model
6.6
Physical
Slot Width
OAM Module
Dual-slot
Length
167 mm 6.6 inches
Height
69 mm 2.7 inches
Outputs
No outputs
4x mini-DisplayPort 2.1
Bus Interface
PCIe 5.0 x16
PCIe 5.0 x8
Other
Launch Price
499 USD
Production
Active
Predecessor
Radeon Instinct
View Instinct MI300X Details View Arc Pro B60 Details