CPU Comparison
Intel Xeon 6728P vs Intel Xeon 6740P
A side-by-side comparison of specs, performance and value. The Intel Xeon 6728P is a 24-core, 48-thread server and workstation processor based on the Granite Rapids-SP platform, offering 144 MB of L3 cache, 8-channel DDR5-6400 memory, and 88 PCIe 5.0 lanes in a 210 W TDP envelope.
The Bottom Line
Overview & Launch
Specifications Compared
Performance Compared
Productivity
Gaming
Virtualization
Efficiency
Specialized Performance
AI / ML
- Intel AMX accelerates INT8 and BF16 matrix operations for deep learning inference.
- Good fit for CPU-based LLM inference and small-to-medium model serving.
- AI performance per core significantly better than pre-AMX Xeon generations.
- Strong CPU-based inference for models that fit in cache and memory.
- No dedicated AI matrix units beyond AVX-512/AMX in this generation.
- Best suited as a host CPU for GPU-accelerated AI training or inference.
Content Creation
Gaming
- No integrated graphics; discrete GPU required.
- High platform cost makes it unattractive for gaming versus consumer CPUs.
- Adequate for casual gaming but not a target use case.
- Not designed or marketed for gaming.
- Lack of integrated graphics and optimized latency for client workloads.
- Better suited for server and enterprise use cases.
Industry Impact
Best CPU by Use Case
Target Audience
Strengths & Weaknesses
Pros
- 24 high-performance Redwood Cove P-cores with strong IPC.
- 144 MB L3 cache benefits latency-sensitive workloads.
- 8-channel DDR5-6400 with up to 4 TB per socket.
- 88 PCIe 5.0 lanes for GPUs, NVMe, and networking.
- AMX and on-die accelerators (QAT, DLB, DSA, IAA) for AI and data processing.
- Supports 2S/4S/8S configurations for scalable enterprise deployments.
Cons
- High platform cost and 210 W TDP require robust cooling and power delivery.
- Locked multiplier; no overclocking headroom.
- No integrated graphics; must pair with discrete GPU or BMC.
- Memory and motherboard ecosystem are more expensive than consumer platforms.
- Less core-count density than higher-end Granite Rapids or EPYC 9004 SKUs.
Pros
- 48 cores and 96 threads for high parallel throughput
- 288 MB shared L3 cache reduces latency for large datasets
- 8-channel DDR5-6400 memory subsystem
- Up to 88 PCIe 5.0 lanes in 2P configs for GPU and NVMe expansion
- Strong platform features (CXL 2.0, Intel AMX, QAT, DSA) for server workloads
- Good performance-per-watt within its core-count and frequency band
Cons
- High 270 W TDP requires robust cooling and power delivery
- Locked multiplier with no overclocking headroom
- No integrated graphics; must be paired with a discreet GPU or BMC
- Platform and memory costs are significant compared to client CPUs
- Single-thread performance is lower than lower-core-count, higher-clocked SKUs
Competitors & Alternatives
Intel Xeon 6728P
- AMD EPYC 9224 (24C/48T, Zen 4, SP5)Rival
Server / Workstation
- Intel Xeon 6521P (24C/48T, Granite Rapids-SP)Rival
Server / Workstation
- Intel Xeon 6731P (32C/64T, Granite Rapids-SP)Rival
Server / Workstation
- AMD EPYC 9124 (16C/32T, Zen 4, SP5)Rival
Server / Workstation
- Intel Xeon Platinum 8380 (40C/80T, Ice Lake-SP)Rival
Server / Workstation
Same 24C/48T Granite Rapids-SP family but lower TDP and price if you don’t need the full 210 W performance profile.
Compare head-to-head- AMD EPYC 9224Alt
24-core Zen 4 server CPU with 12-channel DDR5 and competitive performance; better if you prioritize core count or prefer AMD’s platform.
Higher 32C/64T count within the same Granite Rapids-SP platform if you need more threads and can afford the higher TDP.
Compare head-to-head- AMD EPYC 9124Alt
Lower-cost 16-core Zen 4 server CPU if your workload doesn’t require 24 cores and you want to reduce platform cost.
- Intel Xeon Platinum 8260 (used)Alt
Older 24-core Cascade Lake-SP part available on the used market at lower cost if you don’t need DDR5, PCIe 5.0, or AMX.
Intel Xeon 6740P
- AMD EPYC 9474FRival
Server
- AMD EPYC 9654Rival
Server
- Compare head-to-headIntel Xeon 6760PRival
Server
- Intel Xeon 6730PRival
Server
- Compare head-to-headIntel Xeon 6781PRival
Server
- Intel Xeon 6500P SeriesAlt
Lower-core-count P-core SKUs (e.g., 6530P) with similar platform features but reduced TDP and cost.
Our Verdict on Each
A balanced Granite Rapids-SP SKU with strong per-core performance, large cache, and serious AI acceleration, best suited for memory-intensive and AI-augmented server workloads rather than cost-sensitive or purely throughput-oriented deployments.
Best for: Building or upgrading a 2S/4S server or workstation for AI inference, in-memory databases, or virtualization where 8-channel DDR5 and AMX are valuable.
Read the full reviewA high-core-count Granite Rapids-SP Xeon optimized for 2P servers needing strong memory bandwidth and large cache, though power and platform cost are substantial.
Best for: Building or refreshing a dual-socket server for virtualization, databases, or general enterprise workloads where core density and memory bandwidth matter more than absolute single-thread performance.
Read the full reviewFrequently Asked Questions
Which is better, Intel Xeon 6728P or Intel Xeon 6740P?
Based on our editorial ratings, the Intel Xeon 6740P comes out ahead with a score of 8.7/10. That said, the best choice depends on your workload — check the spec and performance breakdown above for gaming, productivity and efficiency differences.
Which is faster for gaming, Intel Xeon 6728P or Intel Xeon 6740P?
For gaming, the Intel Xeon 6728P leads with a gaming performance score of 60/100 among Intel Xeon 6728P and Intel Xeon 6740P.
Which uses less power?
The Intel Xeon 6728P has the lowest rated TDP. Power draw across these chips: Intel Xeon 6728P (210 W), Intel Xeon 6740P (270 W).
Do Intel Xeon 6728P and Intel Xeon 6740P use the same socket?
Yes — all of these CPUs use the FCLGA4710 socket, so they share compatible motherboards.
Which has more cores?
The Intel Xeon 6740P has the most cores. Core counts: Intel Xeon 6728P (24 cores), Intel Xeon 6740P (48 cores).