AI Accelerator Manufacturing Cost (COGS) — Historical Time Series Data

Estimated manufacturing cost breakdown for NVIDIA A100, H100, H200, B200 and AMD MI300X. Shows HBM shift from 33% to 45%+ of BOM.

AI Accelerator Manufacturing Cost (COGS)

Estimated manufacturing cost breakdown for NVIDIA A100, H100, H200, B200 and AMD MI300X. Shows HBM shift from 33% to 45%+ of BOM.

15 data points·Unit: $·Last updated: Mar 2026

Data Table

PeriodSeriesValue
A100 (2020)
Logic Die
$400
A100 (2020)
HBM
$1,350
A100 (2020)
Packaging & Test
$750
H100 (2023)
Logic Die
$300
H100 (2023)
HBM
$1,350
H100 (2023)
Packaging & Test
$1,670
H200 (2024)
HBM
$1,500
H200 (2024)
Logic Die
$300
H200 (2024)
Packaging & Test
$2,450
MI300X (2024)
Logic Die
$600
MI300X (2024)
HBM
$2,900
MI300X (2024)
Packaging & Test
$1,800
B200 (2025)
Logic Die
$850
B200 (2025)
HBM
$2,900
B200 (2025)
Packaging & Test
$2,650

Methodology & Sources

Sources: Silicon Analysts (H100 Feb 2026), Epoch.ai (B200 Dec 2025), Deep Research (102 sources). NVIDIA/AMD do not disclose COGS; estimates from teardown analysis. Gross margin estimates based on analyst consensus.

Citations:

  • Deep Research (102 sources) (Mar 2026)
  • Silicon Analysts (Feb 2026)
  • Silicon Analysts (Mar 2026)
  • Silicon Analysts / SemiAnalysis (Mar 2026)
  • Epoch.ai (Dec 2025)

Related Tools

Related Analysis

Market intelligence on this data

All market intelligence →
HighPackagingJul 16, 2026

TSMC $265B US CapEx Commitment + ASML 30% EUV Capacity Surge Signal Multi-Year AI Supply Expansion — with Equipment Price Hike Risk

TSMC CapEx: +$100B incremental US commitment (total $265B), 2026 CapEx at high end of $52–56B range; ASML EUV/DUV capacity: +30% per year for 2 years; ASML 2026 revenue guidance raised ~15–19% vs prior midpoint (€38B → €44B midpoint); ASML EUV price hike quantum not yet disclosed — wafer/packaging cost pass-through risk flagged but unquantified.

HighPackagingJul 11, 2026

SK Hynix CEO Flags 2027 as Worst-Ever HBM Supply Crunch; $26.5B IPO Capital Earmarked for Capacity Expansion

Capacity expansion funded: $26.5B raised (new fab + packaging facility + EUV scanners); Supply gap: demand forecasted to exceed supply through 2030+; No specific yield or price-per-unit figure disclosed.

HighMemoryJul 10, 2026

HBM Supply Shortage Persists as SK Hynix Commands 56.4% Market Share; Micron Commits $250B to Domestic DRAM Expansion

HBM shortage: supply constrained with SK Hynix holding 56.4% market share per SEC filing; Micron US DRAM capacity target: 40% of total output domestically by 2035; Micron data center segment gross margin reached 87% on $11.5B revenue (up 103% QoQ), signaling sustained premium pricing on memory inputs to AI hardware.

HighPackagingJul 7, 2026

SK Hynix Commits $706B to HBM/DRAM Scale-Up as ISM Flags Memory Components in Active Short Supply

Capacity: P&T7 packaging facility targeted for end-2027; M17 NAND fab operations targeted H1 2029; SK Group aggregate HBM/DRAM CapEx: ~$706B committed. Supply tightness signal: ISM June 2026 survey flags memory in active short supply (no precise % figure cited). Bernstein projects SK Hynix DRAM gross margins peaking at ~92.7% in Q4 2026, implying sustained elevated HBM ASPs through at least 2028.

Frequently Asked Questions

How much does it cost NVIDIA to manufacture an H100?
The estimated total manufacturing cost (COGS) for an H100 SXM5 is approximately $3,320: roughly $300 for the 814mm² TSMC 4N logic die, $1,350 for HBM3 memory, $750 for CoWoS-S packaging, and $920 for test/assembly.
Why is the B200 logic die cheaper than the H100 despite being more advanced?
The B200 uses a chiplet design with two ~400mm² N4P dies (~800mm² total) instead of a single monolithic 814mm² die like the H100. Each smaller chiplet yields significantly better, driving per-die cost down to roughly $900 total for the logic despite using a more advanced node.
What percentage of AI chip cost is HBM memory?
HBM as a share of total BOM has grown from roughly 14% (A100) to 33% (H100) to 43% (H200) to 45%+ (MI300X). The B200 reverses this trend slightly at ~38% because its chiplet logic die is relatively cheap, but the absolute HBM cost continues to rise.