AMD Instinct MI350P vs Intel Arc Pro B65 Comparison

Head-to-Head Benchmarks

The recorded data shows no direct head-to-head benchmark scores between the AMD Instinct MI350P and the Intel Arc Pro B65. Both products hold a 50th percentile position against all GPUs in the database, with an average benchmark score of zero for each. This indicates that neither part has accumulated comparable workload performance entries at the time of analysis. The absence of scored comparisons means the performance relationship must be derived from their architectural specifications and theoretical throughput figures rather than empirical test results.

The AMD Instinct MI350P delivers 36.04 TFLOPS of FP32 compute and 36.04 TFLOPS of FP16 with a 1:1 ratio. The Intel Arc Pro B65 delivers 12.29 TFLOPS of FP32 and 24.58 TFLOPS of FP16 with a 2:1 ratio. In raw FP32 throughput, the AMD part is approximately 2.93 times higher. In FP16, the AMD part leads by roughly 1.47 times. The Intel part's FP16 advantage over its own FP32 is 2:1, which allows it to reach 24.58 TFLOPS, but that remains below the AMD part's 36.04 TFLOPS in the same precision.

Pixel throughput shows a different relationship. The Intel Arc Pro B65 achieves 192.0 GPixel/s, while the AMD Instinct MI350P records 0 MPixel/s. The AMD part has zero ROPs, so it cannot perform conventional rasterized pixel output. The Intel part has 80 ROPs. Texture throughput also favors AMD: 1,126.4 GTexel/s versus 384.0 GTexel/s for Intel. The AMD part has 512 TMUs, while Intel has 160 TMUs.

Memory bandwidth diverges sharply. The AMD Instinct MI350P reaches 8.19 TB/s over an 8192-bit HBM3e interface, while the Intel Arc Pro B65 reaches 608.0 GB/s over a 256-bit GDDR6 interface. That is a 13.47 times bandwidth advantage for AMD. Memory capacity differs as well: 144 GB versus 32 GB.

Architecture Differences

The AMD Instinct MI350P uses the CDNA 4.0 architecture with the MI350 128CU chip, built on a 3 nm process at TSMC. The Intel Arc Pro B65 uses the Xe2-HPG architecture with the BMG-G21 chip, built on a 5 nm process at TSMC. The process node advantage belongs to AMD, but Intel achieves a higher transistor density: 72.1M per mm² versus 61.3M per mm² for AMD.

Transistor counts separate the two significantly. The AMD chip integrates 73,000 million transistors on a 1190 mm² die. The Intel chip integrates 19,600 million transistors on a 272 mm² die. The AMD die is 4.38 times larger and carries 3.72 times more transistors. The density figures show Intel packs more transistors per area, but the AMD part uses its larger die to deliver substantially more compute and memory resources.

The AMD part has 8192 shading units, 512 TMUs, and zero ROPs. It has no ray tracing cores and no tensor cores listed. The Intel part has 2560 shading units, 160 TMUs, 80 ROPs, and 20 ray tracing cores. The Intel part supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The AMD part reports N/A for DirectX, OpenGL, and Vulkan, reflecting its compute-oriented design with no display outputs. The Intel part has 4x DisplayPort 2.1 outputs.

Clock behavior also differs. The AMD part has a base clock of 1000 MHz and a boost clock of 2200 MHz. The Intel part has a base clock of 2400 MHz and a boost clock of 2400 MHz, meaning it runs at a fixed frequency. Memory clocks differ: AMD runs at 2000 MHz with 8 Gbps effective, while Intel runs at 2375 MHz with 19 Gbps effective. The AMD memory bus width is 8192 bit, versus 256 bit for Intel.

The power envelope is not comparable. The AMD part has a TDP of 600 W with a suggested PSU of 1000 W and a single 16-pin connector. The Intel part has a TDP of 200 W with a suggested PSU of 550 W and a single 8-pin connector. Both are dual-slot cards. The AMD part measures 267 mm in length, 111 mm in height, and 40 mm in width. The Intel part has no recorded dimensions.

Release timing shows the Intel Arc Pro B65 launched on 2026-03-31, while the AMD Instinct MI350P launched on 2026-05-06. The AMD part lists its predecessor as Radeon Instinct, while the Intel part has no predecessor recorded. The Intel part has an active production status, while the AMD part has no production status recorded.

Where Each One Wins

The AMD Instinct MI350P wins decisively in compute throughput scenarios. Its FP32 output of 36.04 TFLOPS is nearly triple the Intel part's 12.29 TFLOPS. Its FP16 output of 36.04 TFLOPS exceeds Intel's 24.58 TFLOPS by about 47 percent. Texture throughput favors AMD by nearly three times, with 1,126.4 GTexel/s versus 384.0 GTexel/s. Memory bandwidth favors AMD by more than an order of magnitude, with 8.19 TB/s versus 608.0 GB/s. The 144 GB memory capacity is 4.5 times larger than Intel's 32 GB.

The Intel Arc Pro B65 wins in conventional graphics and display output scenarios. It has 80 ROPs and delivers 192.0 GPixel/s, while the AMD part has zero ROPs and no pixel output. The Intel part includes 20 ray tracing cores, which the AMD part lacks entirely. The Intel part supports DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, while the AMD part reports no graphics API support. The Intel part provides 4x DisplayPort 2.1 outputs for direct display connection.

The Intel part also wins on efficiency per watt when comparing the raw FP32 to TDP ratio. The Intel part delivers 12.29 TFLOPS at 200 W, yielding 61.45 GFLOPS per watt. The AMD part delivers 36.04 TFLOPS at 600 W, yielding 60.07 GFLOPS per watt. The difference is marginal, with Intel holding a slight 2.3 percent lead in this derived metric. The Intel part also runs at a higher base clock of 2400 MHz versus 1000 MHz for AMD, and its fixed 2400 MHz boost means no clock variability under load.

The AMD part wins on memory clock effective bandwidth per watt. Its 8.19 TB/s at 600 W gives 13.65 GB/s per watt. The Intel part's 608.0 GB/s at 200 W gives 3.04 GB/s per watt. The AMD part is about 4.49 times more efficient in memory bandwidth per watt.

The Verdict

The data clearly separates these two products into different usage domains. The AMD Instinct MI350P is a compute accelerator with no display outputs and no graphics API support. Its design prioritizes massive FP32 and FP16 throughput, enormous memory capacity, and extreme bandwidth. The 8192 shading units, 512 TMUs, 144 GB HBM3e memory, and 8.19 TB/s bandwidth position it for large-scale data processing and AI training workloads where memory capacity and bandwidth are primary constraints.

The Intel Arc Pro B65 is a workstation graphics card with full display output capability and modern graphics API support. Its 80 ROPs, 20 ray tracing cores, and 192.0 GPixel/s pixel rate make it suitable for rendering tasks. The 32 GB GDDR6 memory and 608.0 GB/s bandwidth provide adequate capacity for professional visualization workloads, though they fall far short of the AMD part's memory resources.

The percentile positions are identical at 50, and the average benchmark scores are both zero. That means the database has no empirical evidence to rank one above the other in actual workload performance. The choice depends entirely on the required function. A system needing compute acceleration without display output should use the AMD Instinct MI350P. A system needing graphics rendering with display connectivity should use the Intel Arc Pro B65.

The AMD part requires a 1000 W PSU and a 16-pin connector, while the Intel part requires a 550 W PSU and an 8-pin connector. The AMD part is physically larger at 267 mm length, while the Intel part has no recorded dimensions. The AMD part launched later on 2026-05-06, while the Intel part launched earlier on 2026-03-31. Neither part has a launch MSRP recorded in the database.

FAQ

Q: Which GPU has higher FP32 compute performance?

A: The AMD Instinct MI350P delivers 36.04 TFLOPS of FP32, while the Intel Arc Pro B65 delivers 12.29 TFLOPS. The AMD part is approximately 2.93 times higher.

Q: Does the AMD Instinct MI350P support display outputs?

A: No. The AMD part records "No outputs" for display connections, and its DirectX, OpenGL, and Vulkan support are all listed as N/A. The Intel Arc Pro B65 has 4x DisplayPort 2.1 outputs.

Q: What is the memory capacity difference between the two cards?

A: The AMD Instinct MI350P has 144 GB of HBM3e memory, while the Intel Arc Pro B65 has 32 GB of GDDR6 memory. The AMD part holds 4.5 times more memory.

Q: Which card has ray tracing support?

A: The Intel Arc Pro B65 includes 20 ray tracing cores. The AMD Instinct MI350P has no ray tracing cores listed in the database.

Q: What are the power requirements for each card?

A: The AMD Instinct MI350P has a TDP of 600 W and a suggested PSU of 1000 W with a 16-pin connector. The Intel Arc Pro B65 has a TDP of 200 W and a suggested PSU of 550 W with an 8-pin connector.

Q: Which card has a higher pixel fill rate?

A: The Intel Arc Pro B65 achieves 192.0 GPixel/s with its 80 ROPs. The AMD Instinct MI350P records 0 MPixel/s because it has zero ROPs.