← Taiwan Semiconductor Manufacturing Company Limited
TSM · Forward model · Wei case
The Wei case, 20 quarters out
Each segment is projected from its own operating driver, rolled up into consolidated cash flow, and discounted back to a fair value per share. The assumptions below are editable — change them and every number on this page moves with them.
TSMC has no reportable segments in the Microsoft sense - it is a single foundry business and reports one consolidated profit and loss. What it publishes every quarter, on one slide of its earnings deck, is revenue BY PLATFORM as a percentage of net revenue: High Performance Computing, Smartphone, Internet of Things, Automotive, Digital Consumer Electronics and Others. Those six percentages sum to exactly 100 in all eight quarters here, so multiplying them by the reported dollar revenue gives six lines that reconcile to the consolidated total exactly. That is the whole basis of this model's verticals, and its limitation: the percentages are WHOLE NUMBERS, so every line carries up to half a point of rounding - about $200M a quarter on a $40.2B base, which is half the DCE line and a quarter of the Others line. Read the two small platforms as indicative and the three large ones as sound. What TSMC does NOT publish is any economics by platform. There is no gross margin, no operating margin, no capital expenditure and no capacity by platform - only consolidated figures - so all six verticals here carry the SAME margin and the SAME capital intensity, and the split does no economic work at all. It changes the revenue path and nothing else. The margin used is the disclosed 60.3% operating margin plus depreciation and amortisation added back at 15.14% of revenue, the first-half 2026 rate, giving 75.4%. There is no corporate overhead layer because there is nothing unallocated to put in it. One number in that add-back deserves attention: TSMC's depreciation was NT$359.5bn in the first half of 2026 against NT$359.0bn in the first half of 2025 - essentially FLAT - while capital expenditure ran at NT$496bn in the basis quarter alone. The fabs being built now are not yet in service, so the depreciation from this capital wave has barely started to arrive. Finally, the driver: TSMC publishes wafer shipments only in total (4,336 thousand 12-inch equivalents in the basis quarter) and gives no capacity or shipment figure per platform, so all six verticals run growth drivers. A per-platform wafer driver would have to assume every platform earns the same revenue per wafer, which is the one thing that is certainly false when HPC sits on 2-nanometer and IoT sits on mature nodes. A seventh vertical, Tesla AI5, is carried at ZERO in every reported quarter and starts in 2027. It is not a TSMC disclosure of any kind - TSMC has never named Tesla as a customer, never mentioned AI5 and publishes no wafer price - and it exists on this page to make one question answerable: what is a flagship automotive AI design win worth to a company running at $45bn a quarter? The arithmetic is two million cars a year at two half-reticle dies each, so four million chips or one million a quarter, at roughly $150 of TSMC revenue per die - about $22k a wafer over some 175 good dies, plus packaging. That is $153M in its first quarter, growing to $225M by 2031 as Optimus and training clusters lift the die count toward 1.8 million a quarter. It never exceeds 0.28% of revenue and it moves base fair value by one dollar, from $472.80 to $473.80. Note also that Tesla's EXISTING silicon is already inside the Automotive and HPC lines, which are projected from history that contains it, so this vertical must be read as incremental AI5 volume above the current run rate or it double-counts.
C.C. Wei's version: AI demand is structural rather than cyclical, TSMC's leading-edge monopoly holds through 2nm and A13, and the company keeps both the volume and the price. Every platform is lifted, because the claim is about the foundry rather than one end market. Two things this case does NOT contain. TSMC has published no multi-year revenue target - the furthest forward number in the filings is a single full-year growth rate - so there is no stated claim to test this against, unlike the other founder cases on this site. And at 11x terminal revenue the exit multiple is doing most of the work here; the operating assumptions on their own do not get close. Read it as the price of believing the monopoly persists for five years, not as a model of it.
Latest: $175.10B (2031Q2E)
| Period | Value |
|---|---|
| 2024Q3 | $23.50B |
| 2024Q4 | $26.88B |
| 2025Q1 | $25.53B |
| 2025Q2 | $30.07B |
| 2025Q3 | $33.10B |
| 2025Q4 | $33.73B |
| 2026Q1 | $35.90B |
| 2026Q2 | $40.20B |
| 2026Q3E | $46.00B |
| 2026Q4E | $51.95B |
| 2027Q1E | $58.14B |
| 2027Q2E | $64.20B |
| 2027Q3E | $70.27B |
| 2027Q4E | $76.34B |
| 2028Q1E | $82.42B |
| 2028Q2E | $88.53B |
| 2028Q3E | $94.69B |
| 2028Q4E | $100.94B |
| 2029Q1E | $107.30B |
| 2029Q2E | $113.81B |
| 2029Q3E | $120.50B |
| 2029Q4E | $127.41B |
| 2030Q1E | $134.56B |
| 2030Q2E | $141.99B |
| 2030Q3E | $149.73B |
| 2030Q4E | $157.81B |
| 2031Q1E | $166.26B |
| 2031Q2E | $175.10B |
What drives each segment
High Performance Computing
Growth pathTwo thirds of TSMC and effectively all of its growth. AI accelerators, datacenter CPUs and GPUs, and high-end compute - the platform every AI chip designer in the world routes through. It was 51% of revenue eight quarters ago and 66% now, and it grew 20% sequentially in the basis quarter while the company grew 12%. TSMC does not break out an AI line; AI sits inside this one, and the company will not size it. What makes this platform different from the other five is that it is the one bidding for leading-edge capacity: 2-nanometer began contributing revenue this quarter and 7-nanometer and below is 77% of wafer revenue.
Latest: $153.94B (2031Q2E)
| Period | Value |
|---|---|
| 2024Q3 | $11.98B |
| 2024Q4 | $14.25B |
| 2025Q1 | $15.06B |
| 2025Q2 | $18.04B |
| 2025Q3 | $18.87B |
| 2025Q4 | $18.55B |
| 2026Q1 | $21.90B |
| 2026Q2 | $26.53B |
| 2026Q3E | $31.97B |
| 2026Q4E | $37.58B |
| 2027Q1E | $43.26B |
| 2027Q2E | $48.98B |
| 2027Q3E | $54.71B |
| 2027Q4E | $60.44B |
| 2028Q1E | $66.17B |
| 2028Q2E | $71.94B |
| 2028Q3E | $77.75B |
| 2028Q4E | $83.64B |
| 2029Q1E | $89.65B |
| 2029Q2E | $95.79B |
| 2029Q3E | $102.12B |
| 2029Q4E | $108.65B |
| 2030Q1E | $115.42B |
| 2030Q2E | $122.46B |
| 2030Q3E | $129.80B |
| 2030Q4E | $137.48B |
| 2031Q1E | $145.51B |
| 2031Q2E | $153.94B |
Assumptions & reasoning
- Revenue here is total company revenue multiplied by the platform percentage TSMC publishes on one slide of its quarterly deck. The percentages are whole numbers and sum to exactly 100 in all eight quarters, so the six lines reconcile to reported revenue exactly - but each line carries up to half a percentage point of rounding, which on this platform is about $200M.
- TSMC publishes no revenue, no margin and no capacity for HPC. It publishes a percentage of revenue and a sequential growth rate, and nothing else - which is why this vertical has a growth driver and the same margin as the other five.
- The one volume metric TSMC does publish is company-wide: 4,336 thousand 12-inch equivalent wafers in the basis quarter, up 3.9% sequentially against revenue up 12.0%. The gap is price and mix - blended revenue per wafer rose to about $9,300 - and almost all of it is this platform moving to leading-edge nodes.
- Two-nanometer contributed 3% of wafer revenue for the first time this quarter and 7-nanometer and below reached 77%. That node ramp is the physical event behind this platform's growth, and TSMC guides gross margin DOWN to 65-67% next quarter partly because of it.
Smartphone
Growth pathThe platform HPC is taking share from. $8.8B of revenue, 22% of the company, down 4% sequentially in a quarter when the company grew 12% - and down from 34% of revenue two years ago. It is not shrinking in dollars so much as being outgrown: smartphone revenue is roughly where it was a year ago while HPC has more than doubled. It remains the second-largest platform and the one with the most seasonal shape, since flagship silicon is built in the middle of the calendar year for autumn launches.
Latest: $11.92B (2031Q2E)
| Period | Value |
|---|---|
| 2024Q3 | $7.99B |
| 2024Q4 | $9.41B |
| 2025Q1 | $7.15B |
| 2025Q2 | $8.12B |
| 2025Q3 | $9.93B |
| 2025Q4 | $10.79B |
| 2026Q1 | $9.33B |
| 2026Q2 | $8.84B |
| 2026Q3E | $8.98B |
| 2026Q4E | $9.11B |
| 2027Q1E | $9.25B |
| 2027Q2E | $9.39B |
| 2027Q3E | $9.53B |
| 2027Q4E | $9.67B |
| 2028Q1E | $9.82B |
| 2028Q2E | $9.97B |
| 2028Q3E | $10.12B |
| 2028Q4E | $10.27B |
| 2029Q1E | $10.42B |
| 2029Q2E | $10.58B |
| 2029Q3E | $10.74B |
| 2029Q4E | $10.90B |
| 2030Q1E | $11.07B |
| 2030Q2E | $11.23B |
| 2030Q3E | $11.40B |
| 2030Q4E | $11.57B |
| 2031Q1E | $11.75B |
| 2031Q2E | $11.92B |
Assumptions & reasoning
- Down 4% sequentially in a quarter when the company grew 12%, and down from 34% of revenue to 22% in two years. The dollars are roughly flat year over year; the share loss is HPC growing, not smartphone shrinking.
- This is the most seasonal platform TSMC has - flagship silicon is built mid-year for autumn launches - so a single sequential print is a poor guide. The eight quarters here run +16%, +17%, -22%, +7%, +19%, +11%, -11%, -4%.
- Smartphone competes with HPC for the same leading-edge capacity. When TSMC says demand exceeds supply, this is the platform that gets rationed, which is a reason to model it flat rather than growing.
Internet of Things
Growth path$2.0B, 5% of revenue, up 4% sequentially. Sensors, connectivity and edge silicon, mostly on mature and specialty nodes rather than leading edge. It grew 35% in one quarter two years ago and fell 15% in another, so the quarter-to-quarter path says less about the platform than about a handful of customers' inventory cycles. At 5% of revenue it is carried because TSMC publishes it, not because it decides anything.
Latest: $3.55B (2031Q2E)
| Period | Value |
|---|---|
| 2024Q3 | $1.65B |
| 2024Q4 | $1.34B |
| 2025Q1 | $1.28B |
| 2025Q2 | $1.50B |
| 2025Q3 | $1.66B |
| 2025Q4 | $1.69B |
| 2026Q1 | $2.15B |
| 2026Q2 | $2.01B |
| 2026Q3E | $2.09B |
| 2026Q4E | $2.17B |
| 2027Q1E | $2.24B |
| 2027Q2E | $2.32B |
| 2027Q3E | $2.39B |
| 2027Q4E | $2.46B |
| 2028Q1E | $2.53B |
| 2028Q2E | $2.61B |
| 2028Q3E | $2.68B |
| 2028Q4E | $2.75B |
| 2029Q1E | $2.83B |
| 2029Q2E | $2.90B |
| 2029Q3E | $2.98B |
| 2029Q4E | $3.05B |
| 2030Q1E | $3.13B |
| 2030Q2E | $3.21B |
| 2030Q3E | $3.29B |
| 2030Q4E | $3.38B |
| 2031Q1E | $3.46B |
| 2031Q2E | $3.55B |
Assumptions & reasoning
- Five percent of revenue and the most erratic sequential path of any platform: +35%, -15%, -9%, +14%, +20%, +3%, +12%, +4% across the eight quarters shown. That is customer inventory rather than end demand.
- Mostly mature and specialty nodes, so it does not compete for the leading-edge capacity that constrains HPC and smartphone. It is the platform least exposed to the 2-nanometer ramp in either direction.
Automotive
Growth path$1.6B, 4% of revenue, up 15% sequentially - the second-fastest platform in the basis quarter. ADAS, EV powertrain and in-vehicle compute. Automotive silicon has been through a two-year inventory correction and the sequential prints here have swung from -7% to +18% without much trend. It is the platform with the longest design-in cycles and the least sensitivity to the AI question that decides the rest of this model.
Latest: $3.37B (2031Q2E)
| Period | Value |
|---|---|
| 2024Q3 | $1.18B |
| 2024Q4 | $1.08B |
| 2025Q1 | $1.28B |
| 2025Q2 | $1.50B |
| 2025Q3 | $1.66B |
| 2025Q4 | $1.69B |
| 2026Q1 | $1.44B |
| 2026Q2 | $1.61B |
| 2026Q3E | $1.71B |
| 2026Q4E | $1.80B |
| 2027Q1E | $1.89B |
| 2027Q2E | $1.97B |
| 2027Q3E | $2.06B |
| 2027Q4E | $2.14B |
| 2028Q1E | $2.22B |
| 2028Q2E | $2.30B |
| 2028Q3E | $2.38B |
| 2028Q4E | $2.47B |
| 2029Q1E | $2.55B |
| 2029Q2E | $2.63B |
| 2029Q3E | $2.72B |
| 2029Q4E | $2.80B |
| 2030Q1E | $2.89B |
| 2030Q2E | $2.98B |
| 2030Q3E | $3.08B |
| 2030Q4E | $3.17B |
| 2031Q1E | $3.27B |
| 2031Q2E | $3.37B |
Assumptions & reasoning
- Up 15% sequentially, the second-fastest platform in the basis quarter, but the eight-quarter path is +6%, +6%, +14%, +0%, +18%, -1%, -7%, +15% - a platform coming out of an inventory correction rather than one compounding.
- Automotive design-in cycles run years rather than quarters, so this line is the least sensitive of the six to the AI capacity question that decides the rest of the model.
Digital Consumer Electronics
Growth path$0.4B and 1% of revenue - televisions, set-top boxes and consumer silicon. It is the smallest platform TSMC reports and the one most distorted by rounding: at a whole-percentage disclosure, 1% of a $40.2B quarter could be anything from $200M to $600M. It is carried because TSMC publishes it separately, and the honest thing to say is that its projection cannot be more precise than the disclosure it comes from.
Latest: $628M (2031Q2E)
| Period | Value |
|---|---|
| 2024Q3 | $235M |
| 2024Q4 | $269M |
| 2025Q1 | $255M |
| 2025Q2 | $301M |
| 2025Q3 | $331M |
| 2025Q4 | $337M |
| 2026Q1 | $359M |
| 2026Q2 | $402M |
| 2026Q3E | $414M |
| 2026Q4E | $426M |
| 2027Q1E | $437M |
| 2027Q2E | $448M |
| 2027Q3E | $459M |
| 2027Q4E | $470M |
| 2028Q1E | $480M |
| 2028Q2E | $491M |
| 2028Q3E | $502M |
| 2028Q4E | $513M |
| 2029Q1E | $523M |
| 2029Q2E | $534M |
| 2029Q3E | $546M |
| 2029Q4E | $557M |
| 2030Q1E | $568M |
| 2030Q2E | $580M |
| 2030Q3E | $592M |
| 2030Q4E | $604M |
| 2031Q1E | $616M |
| 2031Q2E | $628M |
Assumptions & reasoning
- One percent of revenue, disclosed as a whole number. The true figure could be anywhere between 0.5% and 1.5%, which is $200M to $600M on a $40.2B quarter - a range wider than the line itself. No projection here can be more precise than that.
- It is carried separately because TSMC reports it separately. Folding it into Others would be tidier and would misstate what the company publishes.
Others
Growth path$0.8B and 2% of revenue - the residual TSMC does not assign to the five named platforms. Like DCE it is a whole-percentage line and carries the same rounding imprecision. It exists here so the six verticals sum to reported revenue rather than to 98% of it.
Latest: $1.42B (2031Q2E)
| Period | Value |
|---|---|
| 2024Q3 | $470M |
| 2024Q4 | $538M |
| 2025Q1 | $511M |
| 2025Q2 | $601M |
| 2025Q3 | $662M |
| 2025Q4 | $675M |
| 2026Q1 | $718M |
| 2026Q2 | $804M |
| 2026Q3E | $836M |
| 2026Q4E | $868M |
| 2027Q1E | $898M |
| 2027Q2E | $927M |
| 2027Q3E | $956M |
| 2027Q4E | $985M |
| 2028Q1E | $1.01B |
| 2028Q2E | $1.04B |
| 2028Q3E | $1.07B |
| 2028Q4E | $1.10B |
| 2029Q1E | $1.13B |
| 2029Q2E | $1.16B |
| 2029Q3E | $1.19B |
| 2029Q4E | $1.22B |
| 2030Q1E | $1.25B |
| 2030Q2E | $1.29B |
| 2030Q3E | $1.32B |
| 2030Q4E | $1.35B |
| 2031Q1E | $1.39B |
| 2031Q2E | $1.42B |
Assumptions & reasoning
- The residual TSMC does not assign to a named platform, at 2% of revenue. It exists in this model so the six verticals sum to reported revenue rather than to 98% of it.
- Like DCE this is a whole-percentage disclosure on a $40B base, so its dollar value carries roughly $200M of rounding either way and its growth rate should not be read closely.
Tesla AI5
Units × priceA line that does not exist yet, carried at zero across every reported quarter and starting in 2027. Tesla's AI5 is a roughly half-reticle inference die that Musk has said will be built at TSMC, two per car, against a stated ambition of two million cars a year. TSMC has never named Tesla as a customer, never mentioned AI5, and publishes no wafer price, so every number on this line is an outside estimate rather than a disclosure. It is here to make one question answerable on the page: what would a flagship automotive AI win actually be worth to a company running at $45bn a quarter? The answer, on the arithmetic below, is about half a percent of revenue - which is itself the finding.
Latest: $274M (2031Q2E)
| Period | Value |
|---|---|
| 2024Q3 | $0.00 |
| 2024Q4 | $0.00 |
| 2025Q1 | $0.00 |
| 2025Q2 | $0.00 |
| 2025Q3 | $0.00 |
| 2025Q4 | $0.00 |
| 2026Q1 | $0.00 |
| 2026Q2 | $0.00 |
| 2026Q3E | $0.00 |
| 2026Q4E | $0.00 |
| 2027Q1E | $158M |
| 2027Q2E | $162M |
| 2027Q3E | $167M |
| 2027Q4E | $173M |
| 2028Q1E | $178M |
| 2028Q2E | $184M |
| 2028Q3E | $190M |
| 2028Q4E | $196M |
| 2029Q1E | $203M |
| 2029Q2E | $210M |
| 2029Q3E | $217M |
| 2029Q4E | $224M |
| 2030Q1E | $232M |
| 2030Q2E | $239M |
| 2030Q3E | $248M |
| 2030Q4E | $256M |
| 2031Q1E | $265M |
| 2031Q2E | $274M |
Assumptions & reasoning
- Zero in every reported quarter, by construction. Tesla already buys silicon from TSMC and that revenue is inside the Automotive and HPC lines already; this vertical is the INCREMENTAL AI5 ramp above the current run rate, and reading it any other way double-counts.
- The arithmetic: two million cars a year at two dies each is four million chips, or one million a quarter. At roughly $150 of TSMC revenue per chip - about $22k a wafer over some 175 good half-reticle dies, plus packaging - that is about $150M a quarter, or $600M a year.
- The chip count grows toward 1.8 million a quarter, which is where Optimus and training clusters would take it. At $150 a chip that is $270M a quarter. A reader who wants the $250M a quarter that circulates for this programme is implicitly assuming roughly six to eight million AI5 dies a year, which is cars plus everything else, not cars alone.
- At its peak this line is about half a percent of TSMC revenue and moves fair value by cents. That is the point of carrying it: a flagship automotive design win at a company this size is a rounding error, and the model says so explicitly rather than leaving it to intuition.
Where each case comes from
Bear case — primary sources
The primary sources this case is built from — filings, calls and posts. Where they are claims by an interested party rather than disclosures, every number in the Bear column is what happens if they are taken at face value.
Q2 2026 press release - the margin guide
First-half 2026 consolidated statements - the depreciation that has not arrived
Bull case — primary sources
The primary sources this case is built from — filings, calls and posts. Where they are claims by an interested party rather than disclosures, every number in the Bull column is what happens if they are taken at face value.
Q2 2026 press release - the node ramp and the full-year guide
- Jul 16, 2026 In the second quarter, shipments of 2-nanometer accounted for 3% of total wafer revenue; 3-nanometer accounted for 30%; 5-nanometer accounted for 33%; and 7-nanometer accounted for 11%. Advanced technologies, defined as 7-nanometer and more advanced technologies, accounted for 77% of total wafer revenue.
- Jul 16, 2026 Revenue is expected to be between US$44.6 billion and US$45.8 billion;
Q2 2026 earnings deck - the annual outlook
Wei case — primary sources
The primary sources this case is built from — filings, calls and posts. Where they are claims by an interested party rather than disclosures, every number in the Wei column is what happens if they are taken at face value.
Q2 2026 press release - what management said about demand
- Jul 16, 2026 Our business in the second quarter was supported by strong demand for our leading-edge process technologies
- Jul 16, 2026 Moving into third quarter 2026, we expect our business to be supported by continued strong demand for our leading-edge process technologies, including the steep ramp-up of our 2-nanometer technology.
Q2 2026 earnings deck - the only forward number TSMC gives
From cash flow to fair value
| Present value of free cash flow, 20 quarters | $612.77B |
| Terminal-year revenue | $648.90B |
| Terminal-year EBITDA | $497.93B |
| Exit multiple, on revenue | 11.0x |
| Terminal value | $7.14T |
| Discounted at 10.0% a year, terminal value becomes | $4.43T |
| Enterprise value | $5.04T |
| Net cash | $78.68B |
| Equity value | $5.12T |
| Shares | 5.19B |
| Fair value per share | $987.87 |
| Against the current price of $427.30 | +131% |
10% on a business with $78.7B of net cash and a 60% operating margin, marked up for a manufacturing base concentrated on one island and an end market - AI - that supplies all of the growth. The exit multiple is the number to argue with, and unusually so here: TSMC publishes no multi-year target, so nineteen of the twenty projected quarters rest on it. 8x terminal revenue on the roughly 72% terminal EBITDA margin in this model is about 11x EBITDA, which is a defensible multiple for a capital-intensive monopoly, against the 12x forward revenue the shares carry today.
Read the other way round: at $427.30 the market is paying 3.8x terminal-year revenue, holding every other assumption on this page fixed. That is the number to argue about.
The projected path
| Quarter | High Performance Computing | Smartphone | Internet of Things | Automotive | Digital Consumer Electronics | Others | Tesla AI5 | Revenue | YoY | EBITDA | Capex | FCF | R40 | PV of FCF |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2026 Q3E | $31.97B | $8.98B | $2.09B | $1.71B | $414M | $836M | $0 | $46.00B | +39% | $36.41B | $17.34B | $15.64B | +73 | $15.27B |
| 2026 Q4E | $37.58B | $9.11B | $2.17B | $1.80B | $426M | $868M | $0 | $51.95B | +54% | $40.99B | $18.98B | $18.05B | +89 | $17.21B |
| 2027 Q1E | $43.26B | $9.25B | $2.24B | $1.89B | $437M | $898M | $158M | $58.14B | +62% | $45.74B | $20.63B | $20.60B | +97 | $19.17B |
| 2027 Q2E | $48.98B | $9.39B | $2.32B | $1.97B | $448M | $927M | $162M | $64.20B | +60% | $50.38B | $22.17B | $23.13B | +96 | $21.03B |
| 2027 Q3E | $54.71B | $9.53B | $2.39B | $2.06B | $459M | $956M | $167M | $70.27B | +53% | $55.00B | $23.67B | $25.69B | +89 | $22.81B |
| 2027 Q4E | $60.44B | $9.67B | $2.46B | $2.14B | $470M | $985M | $173M | $76.34B | +47% | $59.61B | $25.12B | $28.28B | +84 | $24.51B |
| 2028 Q1E | $66.17B | $9.82B | $2.53B | $2.22B | $480M | $1.01B | $178M | $82.42B | +42% | $64.22B | $26.55B | $30.89B | +79 | $26.14B |
| 2028 Q2E | $71.94B | $9.97B | $2.61B | $2.30B | $491M | $1.04B | $184M | $88.53B | +38% | $68.85B | $27.97B | $33.52B | +76 | $27.70B |
| 2028 Q3E | $77.75B | $10.12B | $2.68B | $2.38B | $502M | $1.07B | $190M | $94.69B | +35% | $73.50B | $29.39B | $36.17B | +73 | $29.19B |
| 2028 Q4E | $83.64B | $10.27B | $2.75B | $2.47B | $513M | $1.10B | $196M | $100.94B | +32% | $78.22B | $30.82B | $38.87B | +71 | $30.63B |
| 2029 Q1E | $89.65B | $10.42B | $2.83B | $2.55B | $523M | $1.13B | $203M | $107.30B | +30% | $83.02B | $32.28B | $41.61B | +69 | $32.02B |
| 2029 Q2E | $95.79B | $10.58B | $2.90B | $2.63B | $534M | $1.16B | $210M | $113.81B | +29% | $87.94B | $33.77B | $44.42B | +68 | $33.37B |
| 2029 Q3E | $102.12B | $10.74B | $2.98B | $2.72B | $546M | $1.19B | $217M | $120.50B | +27% | $92.98B | $35.31B | $47.29B | +67 | $34.69B |
| 2029 Q4E | $108.65B | $10.90B | $3.05B | $2.80B | $557M | $1.22B | $224M | $127.41B | +26% | $98.19B | $36.92B | $50.25B | +66 | $36.00B |
| 2030 Q1E | $115.42B | $11.07B | $3.13B | $2.89B | $568M | $1.25B | $232M | $134.56B | +25% | $103.59B | $38.59B | $53.30B | +65 | $37.28B |
| 2030 Q2E | $122.46B | $11.23B | $3.21B | $2.98B | $580M | $1.29B | $239M | $141.99B | +25% | $109.20B | $40.34B | $56.47B | +65 | $38.57B |
| 2030 Q3E | $129.80B | $11.40B | $3.29B | $3.08B | $592M | $1.32B | $248M | $149.73B | +24% | $115.04B | $42.18B | $59.75B | +64 | $39.85B |
| 2030 Q4E | $137.48B | $11.57B | $3.38B | $3.17B | $604M | $1.35B | $256M | $157.81B | +24% | $121.14B | $44.11B | $63.17B | +64 | $41.14B |
| 2031 Q1E | $145.51B | $11.75B | $3.46B | $3.27B | $616M | $1.39B | $265M | $166.26B | +24% | $127.53B | $46.15B | $66.73B | +64 | $42.43B |
| 2031 Q2E | $153.94B | $11.92B | $3.55B | $3.37B | $628M | $1.42B | $274M | $175.10B | +23% | $134.21B | $48.29B | $70.45B | +64 | $43.75B |
Every row is projected. A year-over-year change is shown only where the quarter it compares against exists — an em dash means there is no comparable quarter, not a flat year.
Model revisions
Assumptions are marked to reality as each quarter prints. Every change is appended here, with the fair value the model produced at the time, so the model's own history stays visible.
| Date | Changed | Fair value then | Note |
|---|---|---|---|
| 2026-08-25 | all | — | First build, on the 2026 Q2 basis, from the intake brief at data/models/intake/tsm.json. |
| 2026-08-25 | verticals[teslaAi5], notes | $473.80 | Added a Tesla AI5 vertical with no history, starting 2027 Q1: one million half-reticle dies a quarter at about $150 of TSMC revenue each, growing toward 1.8 million as Optimus and clusters join. Moves base fair value from $472.80 to $473.80. |