AMD Instinct MI308X vs Intel Arc Pro B370 Comparison

AMD
RADEON

AMD Instinct MI308X

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 B370

CORE STATE Panther Lake
VRAM System Shared
CLOCK SPEED 2400 MHz
TDP 25 W
BUS WIDTH System Shared
ARCHITECTURE Xe3-LPG
nm
PROCESS 3 nm
LAUNCH DATE 2026

Analysis: AMD Instinct MI308X vs Intel Arc Pro B370

Head-to-Head Benchmarks

The database contains no recorded benchmark scores for either the AMD Instinct MI308X or the Intel Arc Pro B370. With both parts holding a 50th percentile position among all GPUs in the database and an average benchmark score of zero for each, there is no quantitative performance data to compare. Neither unit registers any wins in head-to-head testing, as the head-to-head benchmark array is empty. This absence of measurements means any performance conclusions must be drawn entirely from the architectural specifications and recorded hardware characteristics rather than from direct application or synthetic test results.

Architecture Differences

The AMD Instinct MI308X and Intel Arc Pro B370 represent fundamentally different design philosophies and deployment targets. The MI308X uses the Aqua Vanjaram chip built on CDNA 3.0 architecture, manufactured by TSMC on a 5 nm process. The die measures 1017 mm² and contains 153,000 million transistors, yielding a transistor density of 150.4 million per square millimeter. This is a massive accelerator designed for compute-intensive workloads, operating as an OAM module with no display outputs and no power connectors, requiring a suggested power supply of 1150 W.

The Intel Arc Pro B370 uses the Panther Lake chip built on Xe3-LPG architecture, manufactured by Intel on a 3 nm process. Its transistor count, die size, and transistor density are all recorded as unknown. This is an integrated graphics processor (IGP) with a slot width of IGP and a bus interface of IGP, meaning it shares system memory rather than having dedicated VRAM. Its display outputs are listed as portable device dependent, indicating it is designed for mobile or integrated deployments where the display connection is handled by the host platform.

The clock behavior differs substantially. The MI308X runs at a base clock of 1000 MHz with a boost of 2100 MHz, while the Arc Pro B370 has a base clock of 300 MHz and a boost of 2400 MHz. Memory configurations are entirely different: the MI308X features 192 GB of HBM3 memory on an 8192-bit bus with 5.32 TB/s bandwidth, operating at 1300 MHz with 5.2 Gbps effective speed. The Arc Pro B370 uses system shared memory, with its bus width, type, and bandwidth all system dependent, meaning its memory performance cannot be specified independently of the host platform.

The compute resources show an order-of-magnitude difference. The MI308X has 19,456 shading units, 1,216 texture mapping units, and zero ROPs, resulting in a pixel rate of 0 MPixel/s and a texture rate of 2,553.6 GTexel/s. Floating-point performance is 81.72 TFLOPS for both FP32 and FP16 with a 1:1 ratio. The Arc Pro B370 has 1,280 shading units, 40 TMUs, 20 ROPs, and 10 ray tracing cores. Its pixel rate is 48.00 GPixel/s, texture rate is 96.00 GTexel/s, FP32 performance is 6.144 TFLOPS, and FP16 performance is 12.29 TFLOPS with a 2:1 ratio. The MI308X thus delivers 13.3 times the FP32 throughput and 6.65 times the FP16 throughput of the Arc Pro B370, though the MI308X has no ray tracing cores while the Arc Pro B370 includes 10.

API support also separates the two. The MI308X lists DirectX, OpenGL, and Vulkan as N/A, reflecting its compute-only design with no graphics output capability. The Arc Pro B370 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, making it a full graphics solution. The power envelope is dramatically different: the MI308X has a TDP of 750 W, while the Arc Pro B370 has a TDP of 25 W, a 30-fold difference. The MI308X was released on December 5, 2023, while the Arc Pro B370 is dated January 26, 2026, with a production status of active for the Intel part and no recorded status for the AMD part.

Where Each One Wins

Given the absence of benchmark data, the wins must be inferred from architectural capabilities. The AMD Instinct MI308X wins decisively in raw compute throughput. Its 81.72 TFLOPS of FP32 performance, 5.32 TB/s memory bandwidth, and 192 GB of HBM3 capacity position it for large-scale scientific computing, AI training, and inference workloads where massive memory and bandwidth are critical. The 8192-bit memory bus and 1216 TMUs support its 2,553.6 GTexel/s texture rate, though the zero ROP count and zero pixel rate indicate it is not intended for any rasterization or graphics output tasks. Its 750 W TDP and OAM module form factor confirm it belongs in server racks with dedicated power delivery.

The Intel Arc Pro B370 wins in integration efficiency and graphics capability. Its 25 W TDP allows deployment in power-constrained environments where the MI308X cannot operate. The 10 ray tracing cores and support for DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4 make it a functional graphics processor, unlike the MI308X which has no display outputs and no graphics API support. The 48.00 GPixel/s pixel rate and 96.00 GTexel/s texture rate, while far below the MI308X in texture throughput, represent actual graphics rendering capability. Its system shared memory architecture eliminates the need for dedicated VRAM, reducing cost and complexity in integrated systems. The 3 nm Intel process and 2400 MHz boost clock indicate a modern, efficient design despite the lower base clock of 300 MHz.

The FP16 performance ratio also distinguishes their intended uses. The MI308X delivers FP16 at a 1:1 ratio with FP32, meaning no dedicated tensor acceleration for reduced precision. The Arc Pro B370 delivers FP16 at a 2:1 ratio, doubling throughput when precision is reduced. This suggests the Intel part may have some advantage in workloads that can tolerate FP16 computation, while the AMD part maintains consistent throughput across precision levels.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The AMD Instinct MI308X delivers 81.72 TFLOPS of FP32 performance, while the Intel Arc Pro B370 provides 6.144 TFLOPS. The MI308X is approximately 13.3 times faster in FP32 throughput.

Q: Does either GPU support ray tracing?

A: Only the Intel Arc Pro B370 includes ray tracing cores, with 10 RT cores recorded. The AMD Instinct MI308X has no ray tracing cores listed in its specifications.

Q: What memory configurations do these GPUs use?

A: The AMD Instinct MI308X uses 192 GB of HBM3 memory on an 8192-bit bus with 5.32 TB/s bandwidth. The Intel Arc Pro B370 uses system shared memory, with its type, bus width, and bandwidth all dependent on the host system.

Q: What are the power requirements for each GPU?

A: The AMD Instinct MI308X has a TDP of 750 W and requires a suggested power supply of 1150 W. The Intel Arc Pro B370 has a TDP of 25 W and has no suggested PSU listed, consistent with its IGP slot width.

Q: Can either GPU be used for graphics output?

A: The AMD Instinct MI308X has no display outputs and lists DirectX, OpenGL, and Vulkan as N/A. The Intel Arc Pro B370 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, with display outputs that are portable device dependent.

Q: What are the release dates for these products?

A: The AMD Instinct MI308X was released on December 5, 2023. The Intel Arc Pro B370 has a release date of January 26, 2026, and its production status is recorded as active.

The Verdict

The data presents two GPUs with no overlapping use cases. The AMD Instinct MI308X is a compute accelerator with 19,456 shading units, 192 GB of HBM3 memory, 5.32 TB/s bandwidth, and 81.72 TFLOPS of FP32 throughput. It has no graphics capabilities whatsoever, no display outputs, no graphics API support, and no ROPs. Its 750 W TDP and OAM module form factor require dedicated server infrastructure. Any workload involving rendering, display output, or graphics API calls cannot use this device.

The Intel Arc Pro B370 is an integrated graphics processor with 1,280 shading units, 40 TMUs, 20 ROPs, and 10 ray tracing cores. It supports the full modern graphics API stack including DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. Its 25 W TDP and IGP slot width make it suitable for mobile or embedded platforms where power efficiency and integration are priorities. Its 6.144 TFLOPS FP32 performance is a fraction of the MI308X, but it delivers actual graphics functionality that the MI308X cannot provide.

The choice between these two is determined entirely by workload type. For large-scale compute tasks requiring massive memory capacity, extreme bandwidth, and high floating-point throughput, the MI308X is the only viable option. For graphics rendering, ray tracing, and API-compatible applications in power-constrained systems, the Arc Pro B370 is the appropriate selection. Neither can substitute for the other, and the 50th percentile ranking for both in the database reflects their equal standing within their respective, entirely different categories rather than any equivalence in performance.

Specification Differences

| Specification | AMD Instinct MI308X | Intel Arc Pro B370 |

|---|---|---|

| Architecture | CDNA 3.0 | Xe3-LPG |

| Process Node | 5 nm | 3 nm |

| Foundry | TSMC | Intel |

| Transistors | 153,000 million | unknown |

| Die Size | 1017 mm² | unknown |

| Transistor Density | 150.4M / mm² | null |

| Base Clock | 1000 MHz | 300 MHz |

| Boost Clock | 2100 MHz | 2400 MHz |

| Memory Size | 192 GB | System Shared |

| Memory Type | HBM3 | System Shared |

| Memory Bus Width | 8192 bit | System Shared |

| Memory Bandwidth | 5.32 TB/s | System Dependent |

| Shading Units | 19456 | 1280 |

| TMUs | 1216 | 40 |

| ROPs | 0 | 20 |

| Ray Tracing Cores | null | 10 |

| Pixel Rate | 0 MPixel/s | 48.00 GPixel/s |

| Texture Rate | 2,553.6 GTexel/s | 96.00 GTexel/s |

| FP32 Performance | 81.72 TFLOPS | 6.144 TFLOPS |

| FP16 Performance | 81.72 TFLOPS (1:1) | 12.29 TFLOPS (2:1) |

| TDP | 750 W | 25 W |

| Slot Width | OAM Module | IGP |

| Power Connectors | None | None |

| Suggested PSU | 1150 W | null |

| Bus Interface | PCIe 5.0 x16 | IGP |

| Display Outputs | No outputs | Portable Device Dependent |

| DirectX Support | N/A | 12 Ultimate (12_2) |

| OpenGL Support | N/A | 4.6 |

| Vulkan Support | N/A | 1.4 |

| Release Date | 2023-12-05 | 2026-01-26 |

| Production Status | null | Active |

| Predecessor | Radeon Instinct | HD Graphics-WM |

DETAILED SPECIFICATIONS

SPECIFICATION
Instinct MI308X
Pro B370
Core Specs
Shading Units
19,456
1,280 -93.4%
Shaders
19,456
1,280 -93.4%
TMUs
1,216
40 -96.7%
ROPs
0
20 +∞%
Compute Units
304
Execution Units
10
Clocks
Base Clock
1000 MHz
300 MHz
Boost Clock
2100 MHz
2400 MHz
Memory Clock
1300 MHz 5.2 Gbps effective
System Shared
Memory
Memory Size
192 GB
System Shared
VRAM (MB)
196,608
Memory Type
HBM3
System Shared
Memory Bus
8192 bit
System Shared
Bandwidth
5.32 TB/s
System Dependent
Cache
L1 Cache
16 KB (per CU)
64 KB (per EU)
L2 Cache
16 MB
16 MB
L3 Cache
256 MB
Performance
Pixel Rate
0 MPixel/s
48.00 GPixel/s
Texture Rate
2,553.6 GTexel/s
96.00 GTexel/s
FP32 (TFLOPS)
81.72 TFLOPS
6.144 TFLOPS
FP64 (TFLOPS)
40.86 TFLOPS (1:2)
768.0 GFLOPS (1:8)
FP16 (TFLOPS)
81.72 TFLOPS (1:1)
12.29 TFLOPS (2:1)
AI/RT
RT Cores
10
XMX Cores
80
Matrix Cores
1,216
Power
TDP
750 W
25 W
TDP (W)
750
25 -96.7%
Suggested PSU
1150 W
Power Connectors
None
None
Architecture
Architecture
CDNA 3.0
Xe3-LPG
GPU Name
Aqua Vanjaram
Panther Lake
Generation
Instinct (MIx)
Arc Graphics-WM (Panther Lake)
Process Size
5 nm
3 nm
Transistors
153,000 million
unknown
Die Size
1017 mm²
unknown
Foundry
TSMC
Intel
Density
150.4M / 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
6.9
Physical
Slot Width
OAM Module
IGP
Outputs
No outputs
Portable Device Dependent
Bus Interface
PCIe 5.0 x16
IGP
Other
Production
Active
Predecessor
Radeon Instinct
HD Graphics-WM
View Instinct MI308X Details View Arc Pro B370 Details