Intel Arc B370 vs NVIDIA B300 Comparison

Intel
GPU

Intel Arc 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
VS
NVIDIA
GEFORCE

B300

CORE STATE GB110
VRAM 144 GB
CLOCK SPEED 2032 MHz
TDP 1400 W
BUS WIDTH 4096 bit
ARCHITECTURE Blackwell Ultra
nm
PROCESS 5 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

3dmark_3dmark_steel_nomad_dx12
1,184
N/A

Analysis: Intel Arc B370 vs NVIDIA B300

Head-to-Head Benchmarks

The recorded data contains a single benchmark result for the Intel Arc B370: a 3DMark Steel Nomad DX12 score of 1184. The NVIDIA B300 has no recorded benchmark scores in the database, so direct head-to-head comparisons rely on the Arc B370's position against its nearest rivals and the B300's percentile placement.

The Arc B370's score of 1184 places it in the 5th percentile of all GPUs. Its nearest rivals are tightly clustered: the ATI Mobility Radeon HD 5570 scores 1186, which is 0.2% higher than the Arc B370; the ATI Radeon HD 5770 scores 1190, a 0.5% advantage; the AMD Radeon HD 7650M scores 1192, 0.7% ahead. On the other side, the AMD FirePro M2000 scores 1168, which is 1.4% lower than the Arc B370. These deltas are small, indicating the Arc B370 sits in a narrow performance band where movement of a few points changes its standing.

The NVIDIA B300 sits in the 50th percentile of all GPUs, a substantially higher placement than the Arc B370's 5th percentile. However, the B300 has no average benchmark score in the database, and no rivals are listed. The 50th percentile placement suggests the B300 is not a top-tier performer across the full GPU database, but the lack of a concrete score prevents a numeric comparison.

The data shows no head-to-head benchmark entries between the two items. The Arc B370 holds a recorded score; the B300 does not. Any performance relationship between them must be inferred from their percentile positions, which indicate a wide gap, though the absence of a B300 score limits precision.

Architecture Differences

The Intel Arc B370 is built on the Panther Lake chip using the Xe3-LPG architecture, part of the Arc Graphics-M (Panther Lake) generation. It uses a 3 nm process from Intel. The NVIDIA B300 uses the GB110 chip with the Blackwell Ultra architecture, part of the Server Blackwell (Bxx) generation, fabricated on a 5 nm process from TSMC. The node difference is significant: 3 nm versus 5 nm, with Intel's process being the more advanced of the two.

The B300 contains 104,000 million transistors, while the Arc B370's transistor count is listed as unknown. The B300's die size is also not recorded, nor is the Arc B370's. The B300 is a server-class part with a 1400 W TDP, whereas the Arc B370 is an integrated graphics processor (IGP) with a 25 W TDP. The form factors reflect this: the Arc B370 is an IGP with no power connectors, while the B300 is an SXM Module with a suggested PSU of 1800 W.

Memory configurations differ completely. The Arc B370 uses system-shared memory, with type, bus width, and bandwidth all listed as system-dependent. The B300 has 144 GB of HBM3e memory on a 4096-bit bus, delivering 4.10 TB/s of bandwidth. The B300's memory clock is 2000 MHz with 8 Gbps effective, while the Arc B370's memory clock is tied to the system.

The Arc B370 supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The B300 lists no API support in the database. Display outputs differ as well: the Arc B370's outputs are portable-device dependent, while the B300 has no outputs. The bus interface for the Arc B370 is IGP, whereas the B300 uses PCIe 5.0 x16.

FAQ

Q: Which GPU has a higher benchmark score in the database?

A: The Intel Arc B370 has a recorded 3DMark Steel Nomad DX12 score of 1184. The NVIDIA B300 has no recorded benchmark scores, so no direct score comparison is possible.

Q: How does the Arc B370 compare to its nearest rivals?

A: The Arc B370 trails the ATI Mobility Radeon HD 5570 by 0.2%, the ATI Radeon HD 5770 by 0.5%, and the AMD Radeon HD 7650M by 0.7%. It leads the AMD FirePro M2000 by 1.4%.

Q: What are the memory specifications of each GPU?

A: The Arc B370 uses system-shared memory with system-dependent bandwidth. The B300 has 144 GB of HBM3e memory on a 4096-bit bus with 4.10 TB/s bandwidth.

Q: What process nodes do the two GPUs use?

A: The Arc B370 uses a 3 nm process from Intel. The B300 uses a 5 nm process from TSMC.

Q: What is the TDP difference between the two?

A: The Arc B370 has a 25 W TDP and is an IGP. The B300 has a 1400 W TDP and is an SXM Module with a suggested PSU of 1800 W.

Q: Which GPU has more shading units?

A: The B300 has 18,944 shading units, while the Arc B370 has 1,280. The B300 also has 592 tensor cores and 592 TMUs, against the Arc B370's 40 TMUs and no listed tensor cores.

Specification Differences

The two GPUs differ across nearly every recorded specification.

  • Chip: Intel Arc B370 uses Panther Lake; NVIDIA B300 uses GB110.
  • Architecture: Xe3-LPG for the Arc B370; Blackwell Ultra for the B300.
  • Generation: Arc Graphics-M (Panther Lake) for the Arc B370; Server Blackwell (Bxx) for the B300.
  • Process node: 3 nm (Intel) for the Arc B370; 5 nm (TSMC) for the B300.
  • Transistors: Unknown for the Arc B370; 104,000 million for the B300.
  • Base clock: 300 MHz for the Arc B370; 1665 MHz for the B300.
  • Boost clock: 2400 MHz for the Arc B370; 2032 MHz for the B300.
  • Memory clock: System shared for the Arc B370; 2000 MHz with 8 Gbps effective for the B300.
  • Memory size: System shared for the Arc B370; 144 GB for the B300.
  • Memory type: System shared for the Arc B370; HBM3e for the B300.
  • Memory bus width: System shared for the Arc B370; 4096 bit for the B300.
  • Memory bandwidth: System dependent for the Arc B370; 4.10 TB/s for the B300.
  • Shading units: 1280 for the Arc B370; 18,944 for the B300.
  • TMUs: 40 for the Arc B370; 592 for the B300.
  • ROPs: 20 for the Arc B370; 24 for the B300.
  • RT cores: 10 for the Arc B370; none listed for the B300.
  • Tensor cores: None listed for the Arc B370; 592 for the B300.
  • Pixel rate: 48.00 GPixel/s for the Arc B370; 48.77 GPixel/s for the B300.
  • Texture rate: 96.00 GTexel/s for the Arc B370; 1,202.9 GTexel/s for the B300.
  • FP32 performance: 6.144 TFLOPS for the Arc B370; 76.99 TFLOPS for the B300.
  • FP16 performance: 12.29 TFLOPS (2:1) for the Arc B370; 1,231.8 TFLOPS (16:1) for the B300.
  • TDP: 25 W for the Arc B370; 1400 W for the B300.
  • Slot width: IGP for the Arc B370; SXM Module for the B300.
  • Power connectors: None for the Arc B370; not listed for the B300.
  • Suggested PSU: Not listed for the Arc B370; 1800 W for the B300.
  • Bus interface: IGP for the Arc B370; PCIe 5.0 x16 for the B300.
  • Display outputs: Portable device dependent for the Arc B370; no outputs for the B300.
  • APIs: DirectX 12 Ultimate (12_2), OpenGL 4.6, Vulkan 1.4 for the Arc B370; none listed for the B300.
  • Release date: 2026-01-26 for the Arc B370; 2025-09-10 for the B300.
  • Predecessor: None listed for the Arc B370; Server Hopper for the B300.
  • Successor: None listed for the Arc B370; Server Rubin for the B300.

Where Each One Wins

The Arc B370 wins on integration and portability. It is an IGP with a 25 W TDP, no power connectors, and a bus interface of IGP, making it suitable for compact systems where a discrete card is not an option. Its display outputs are dependent on the portable device, which fits a mobile or integrated context. It also brings a full API stack, including DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, which the B300 lacks entirely in the database. Its 10 RT cores provide hardware ray tracing support, a feature the B300 does not list.

The B300 wins on raw compute and memory capacity. Its 18,944 shading units, 592 tensor cores, and 592 TMUs dwarf the Arc B370's 1,280 shading units and 40 TMUs. FP32 performance is 76.99 TFLOPS versus 6.144 TFLOPS, and FP16 performance reaches 1,231.8 TFLOPS against 12.29 TFLOPS. The 144 GB HBM3e memory with 4.10 TB/s bandwidth is a server-class configuration, while the Arc B370 relies on system-shared memory with dependent bandwidth. The B300 also has a higher pixel rate (48.77 GPixel/s versus 48.00 GPixel/s) and a dramatically higher texture rate (1,202.9 GTexel/s versus 96.00 GTexel/s).

The benchmark data shows the Arc B370 with a concrete score of 1184 in 3DMark Steel Nomad DX12, placing it in the 5th percentile. The B300 has no benchmark scores recorded, so its 50th percentile placement is not supported by a measured result in the database. The Arc B370's nearest rivals are all within 1.4% of its score, indicating a tight competitive cluster at its performance level. The B300's position in the 50th percentile suggests a different performance tier, but without a recorded score, the database does not quantify the gap.

The use cases split along architectural lines. The Arc B370 suits systems needing integrated graphics with modern API coverage and ray tracing support at low power. The B300 targets server workloads requiring massive memory capacity, high-bandwidth HBM3e, and tensor core acceleration, with a power envelope that demands an 1800 W suggested PSU. The B300's predecessor is Server Hopper and its successor is Server Rubin, confirming its placement in a server product line, while the Arc B370 belongs to the Arc Graphics-M (Panther Lake) generation for mobile and integrated use.

DETAILED SPECIFICATIONS

SPECIFICATION
B370
B300
Core Specs
Shading Units
1,280
18,944 +1380.0%
Shaders
1,280
18,944 +1380.0%
TMUs
40
592 +1380.0%
ROPs
20
24 +20.0%
SM Count
148
Execution Units
10
Clocks
Base Clock
300 MHz
1665 MHz
Boost Clock
2400 MHz
2032 MHz
Memory Clock
System Shared
2000 MHz 8 Gbps effective
Memory
Memory Size
System Shared
144 GB
VRAM (MB)
147,456
Memory Type
System Shared
HBM3e
Memory Bus
System Shared
4096 bit
Bandwidth
System Dependent
4.10 TB/s
Cache
L1 Cache
64 KB (per EU)
256 KB (per SM)
L2 Cache
16 MB
50 MB
Performance
Pixel Rate
48.00 GPixel/s
48.77 GPixel/s
Texture Rate
96.00 GTexel/s
1,202.9 GTexel/s
FP32 (TFLOPS)
6.144 TFLOPS
76.99 TFLOPS
FP64 (TFLOPS)
768.0 GFLOPS (1:8)
1,202.9 GFLOPS (1:64)
FP16 (TFLOPS)
12.29 TFLOPS (2:1)
1,231.8 TFLOPS (16:1)
AI/RT
RT Cores
10
Tensor Cores
592
XMX Cores
80
Power
TDP
25 W
1400 W
TDP (W)
25
1,400 +5500.0%
Suggested PSU
1800 W
Power Connectors
None
Architecture
Architecture
Xe3-LPG
Blackwell Ultra
GPU Name
Panther Lake
GB110
Generation
Arc Graphics-M (Panther Lake)
Server Blackwell (Bxx)
Process Size
3 nm
5 nm
Transistors
unknown
104,000 million
Die Size
unknown
Foundry
Intel
TSMC
API Support
DirectX
12 Ultimate (12_2)
OpenGL
4.6
Vulkan
1.4
OpenCL
3.0
3.0
CUDA
10.3
Shader Model
6.9
Physical
Slot Width
IGP
SXM Module
Outputs
Portable Device Dependent
No outputs
Bus Interface
IGP
PCIe 5.0 x16
Other
Production
Active
Active
Predecessor
Server Hopper
Successor
Server Rubin
View Arc B370 Details View B300 Details