# Sivers Semiconductors: the opportunity is real; the revenue ramp still has to arrive

Research date: **20 October 2026**. Company: **Sivers Semiconductors AB**. Listing: **Nasdaq Stockholm, SIVE**. ISIN: **SE0003917798**. Financial reporting currency: **SEK**. Latest financial period reviewed: **six months ended 30 June 2026**. Material announcements reviewed through **20 October 2026**.

## Idea origination: why we are researching Serenity's SIVE thesis

**This idea came onto our research agenda through Serenity (@aleabitoreddit), a prominent voice in online AI-photonics investing discussions.** The reader who requested this investigation highlighted his repeated interest in Sivers. We are making that discovery trail explicit: [Serenity's public X account](https://x.com/aleabitoreddit), the independent [Serenity Tracker](https://serenitytrades.com/), and its [Sivers thesis reconstruction](https://serenitytrades.com/deep-dive/sive).

The tracker characterizes SIVE as a high-conviction expression of an upstream laser-supply bottleneck in AI optical connectivity. That is an interesting **investment hypothesis**: a specialist component supplier might capture more of the economics as customers scale. The question for this memo is how much of that hypothesis can be underwritten with disclosed commercial progress, financial results and a defensible valuation. Here, **underwriting** means deciding which assumptions the evidence can reasonably support.

Serenity receives credit for the idea being investigated. Company filings establish the reported financial baseline. Our memo contributes the evidence checks, competing interpretations and outstanding diligence questions. These are different roles in the research process; none requires hiding the original source.

The tracker is a derived, independent reconstruction, not Serenity's original words or a verified brokerage statement. We link the [28 August Lookonchain summary](https://lookonchain.com/feeds/70304) and the [original X post it references](https://x.com/aleabitoreddit/status/2093014569279189295) as an additional trail. The original post body was not directly retrievable during this review, so attributed views below remain secondary-source summaries. We do not assert an audited performance record, exact recommendation rank, present position size or endorsement of this website.

## Investment memo: our current view

**Commercialization watchlist; valuation work remains open.** Sivers has products, disclosed customer programs and an emerging production-order story. The 20 October 2026 LiDAR production orders (shipments from Q4 2026, production in 2027, USD 53 to 138 million cumulative lifecycle potential) advance one program beyond a design win and are the first delivered milestone of the Q2 2026 letter. That is a meaningful milestone, not yet recognised revenue. The next test is whether these programs generate enough recurring product revenue and cash to support the manufacturing investment.

The bullish narrative runs ahead of the current accounts: first-half sales fell while operating cash outflow increased. An expanding sales pipeline can coexist with weak reported results. Neither observation alone settles the investment case.

This first report establishes the operating evidence and an explicit valuation framework. It does **not** establish that today's shares are cheap: an independently verified current quote and an up-to-date reconciliation of cash, shares and dilution are not yet in this review. There is no price target or expected-return forecast.

## Thesis underwriting: from source idea to investable conclusion

The source thesis gives us a set of questions to investigate. It does not supply the answer to all of them. This is our analytical framework, rather than a claim about Serenity's exact language or current portfolio.

| Underwriting question | The bullish proposition | What a disciplined investor needs to establish | Our current assessment |
| --- | --- | --- | --- |
| **Industry structure and scarcity value** | A hard-to-replace laser component becomes a bottleneck as optical links scale | Substitution risk, qualified competing suppliers, yield, customer dependence and pricing power | Product relevance is supported; durable scarcity and excess returns have not been demonstrated in this memo |
| **Commercial conversion** | Engineering engagement becomes recurring production demand | Qualification milestones, firm order size, delivery schedules and repeat purchases | Today's LiDAR orders advance one program; wider customer ramps remain to be measured |
| **Normalized earnings power** | Larger product volumes create operating leverage | Revenue by product, gross margin, fixed-cost absorption, working capital and reinvestment | The operating model is still loss-making; future margin expansion is a hypothesis |
| **Capital allocation and return on invested capital** | Manufacturing investment earns attractive long-term returns | Capex phasing, utilization, cash generation and dilution-adjusted value per share | Capacity plans are disclosed; investment returns cannot yet be underwritten |
| **Variant perception** | Our view differs materially from expectations embedded in the share price | A current valuation, a credible expectations baseline and evidence for a different outcome | Not established: this memo has not yet measured what today's market price implies |
| **Catalyst path and rerating** | Results, customer ramps or a potential US listing change investor expectations | A dated sequence of observable events and a reason they change cash-flow expectations or required returns | We can track milestones; a listing or publicity alone does not establish higher intrinsic value |
| **Risk/reward and margin of safety** | Conservative realizations still leave attractive upside against a defined downside | Bear/base/bull operating assumptions, financing needs, dilution and valuation sensitivity | Pending current market inputs; no numerical upside/downside ratio is asserted |

**The central debate:** does Sivers become an economically attractive volume supplier, or does it remain a technically relevant business that requires substantial capital before shareholders see cash? This is the bridge between a compelling supply-chain narrative and an investment conclusion.

**Our burden of proof:** recurring deliveries and product economics must substantiate the growth story; valuation must then leave adequate compensation for delay, financing and competition. An insightful source can generate an excellent research question even when the share price already discounts a favorable answer.

## What the business actually sells

Sivers has two businesses. **Photonics** supplies semiconductor lasers and optical components; its investment narrative includes the light sources used in optical data links and LiDAR sensing. **Wireless** supplies radio-frequency components such as beamforming chips for satellite communications and wireless connectivity. These have different customers, development cycles and commercialization risks. The [company's business overview](https://www.sivers-semiconductors.com/) describes both.

A useful distinction is between selling engineering work, winning a place in a customer's design, receiving a production order and earning revenue on delivered products. Our interpretation: the investment case becomes stronger as evidence moves through that sequence, particularly when deliveries lead to repeat orders and cash collections. Counting every partnership as production revenue would skip the difficult part.

## The financial starting point

Amounts below are **SEK million**, unless stated otherwise. Comparatives are those presented in the [Q2 2026 interim report, pages 2, 15–16](https://www.sivers-semiconductors.com/wp-content/uploads/2026/08/Sivers-Interim-report-Q226.pdf). Columns compare identical periods; they are not quarterly run-rate forecasts.

| Measure | Q2 2026 | Q2 2025 | H1 2026 | H1 2025 |
| --- | ---: | ---: | ---: | ---: |
| Revenue | 53.8 | 61.4 | 115.6 | 144.3 |
| Wireless revenue | 37.9 | 44.2 | 81.9 | 100.5 |
| Photonics revenue | 15.9 | 17.2 | 33.7 | 43.9 |
| Hardware revenue | 29.2 | 25.7 | 56.8 | 66.9 |
| Non-recurring engineering revenue | 24.6 | 35.5 | 58.8 | 77.1 |
| Adjusted EBITDA | -35.5 | -20.9 | -49.3 | -12.6 |
| Operating cash flow | -70.0 | -20.1 | -119.2 | -37.5 |

The [results announcement](https://www.sivers-semiconductors.com/press/sivers-semiconductors-reports-q2-2026-results-as-product-growth-record-pipeline-and-customer-ramps-position-company-for-growth-acceleration/) distinguishes Q2 product growth of **13% as reported** from **18% at constant exchange rates**. Neither means total revenue grew: reported total sales fell 12%. The same release reports a SEK 42.9 million non-cash social-security accounting expense linked to the share-price rise. Consequently, adjusted EBITDA, reported earnings and cash flow answer different questions; none should silently replace the others.

Our calculations from the table: hardware represented about **54% of Q2 sales** (29.2 / 53.8), but about **49% of first-half sales** (56.8 / 115.6). The latest quarter's mix is encouraging; the six-month figures do not yet demonstrate sustained product expansion.

## What changed today

On **20 October 2026**, Sivers announced first production orders from its strategic LiDAR customer. Shipments are expected to start in Q4 2026, with automotive and industrial programs entering production during 2027. The disclosed **USD 53–138 million** is estimated cumulative revenue potential over the customer's product lifecycle. It is not the value of the first order or a one-year sales forecast. The release does not state the initial order value or identify the customer. [LiDAR production announcement](https://www.sivers-semiconductors.com/press/strategic-lidar-customer-initiates-production-orders-with-sivers-semiconductors/)

Our assessment: this improves the evidence that the program can commercialize. Size, shipment cadence, margins and cash conversion remain to be demonstrated. We do not infer a customer's name from social discussion.

## Pipeline, collaboration, orders and revenue

| Item | What the evidence supports | What it does not establish |
| --- | --- | --- |
| USD 1.2 billion opportunity pipeline | Management's estimate of potential customer business | Guaranteed backlog, annual revenue or cash receipts |
| GlobalFoundries collaboration | Laser arrays intended for reference designs and optical-engine solutions | A disclosed minimum purchase commitment |
| LiDAR production orders | One program progressing into production ordering | USD 53–138 million of firm orders already booked |
| Glasgow capacity expansion | A manufacturing investment intended to support future demand | Full utilization, achieved yield or a proven return on that investment |

The pipeline's own footnote defines it as **non-binding estimated opportunities over 2026–2030**, subject to customer decisions and timing. [Interim report, page 3](https://www.sivers-semiconductors.com/wp-content/uploads/2026/08/Sivers-Interim-report-Q226.pdf)

The [GlobalFoundries announcement](https://www.sivers-semiconductors.com/press/sivers-globalfoundries-advance-ai-data-center-optical-solutions/) describes reference-design integration across optical architectures. It provides technical and commercial context, without disclosing a minimum committed order value. Our interpretation: an integration route is valuable, but qualification, customer selection and shipment economics remain separate hurdles.

## The strongest bull case

Sivers could occupy a valuable component position as customers increase optical connectivity and sensing deployment. Its opportunity extends across several architectures rather than depending exclusively on one co-packaged-optics configuration. Existing technical work could become repeat product revenue, allowing development and fixed manufacturing costs to be spread across greater shipment volumes.

The [September manufacturing announcement](https://www.sivers-semiconductors.com/press/sivers-semiconductors-invests-usd-30-million-to-expand-european-photonics-manufacturing-for-ai-datacenters/) sets out a **USD 30 million** Glasgow expansion, with capacity intended to exceed **100 million CW DFB lasers annually** and operations expected in **Q4 2027**. Our interpretation: if demand and manufacturing execution both materialize, capacity could become a competitive advantage. Those conditions are essential; announced capacity is not sold output.

## The strongest bear case

The company must turn engineering and customer interest into substantial profitable shipments while spending ahead of them. A delay can hurt twice: expected sales arrive later while staffing, facilities and development still consume cash. Even an attractive end market does not determine one supplier's market share, pricing power or eventual return on capital.

Capacity economics add execution risk. Yield, utilization, customer concentration and working-capital requirements matter alongside nameplate capacity. Shareholders also need to follow per-share outcomes: successful fundraising can improve survival prospects while diluting existing ownership.

These are our scenario risks, not allegations of undisclosed events. The observed negative operating cash flow and announced manufacturing commitment explain why we prioritize these questions.

## Financing: avoid combining different dates

The issuer's [Q2 results announcement](https://www.sivers-semiconductors.com/press/sivers-semiconductors-reports-q2-2026-results-as-product-growth-record-pipeline-and-customer-ramps-position-company-for-growth-acceleration/) describes approximately **SEK 825 million** of gross directed issues during the quarter, including the large issue resolved at quarter end, and subsequent conversion of the **USD 12 million** loan. A resolution, settlement, debt conversion and cash receipt are distinct events. Gross proceeds also differ from usable net cash.

The [CEO's August letter](https://www.sivers-semiconductors.com/2026/08/27/vickram-vathulyas-letter-to-shareholders-interim-report-q2-2026/) says the remaining USD 5 million term loan was repaid in August. Treat that as a dated management statement; it is not a verified 20 October cash balance or proof that every lease or other obligation disappeared.

We do not compute a current runway or enterprise value from the June cash balance plus headline fundraising amounts. The missing bridge is: opening cash, settled net proceeds, cash spent, loan repayments, subsequent working capital and remaining obligations.

## Valuation: a framework with explicit missing inputs

The [29 September EGM notice](https://www.sivers-semiconductors.com/press/notice-to-attend-an-extraordinary-general-meeting-of-sivers-semiconductors-ab/) states **356,740,332 issued ordinary shares**, including **12,872,916 held by the company**. That implies **343,867,416 ordinary shares outside treasury** on that date. This calculation is not a fully diluted forecast or confirmation of the count today.

The Q2 report presents FY2025 sales of **SEK 302.8 million**. The following is **our illustrative arithmetic**, using September shares outside treasury and that historical sales figure. Prices are hypothetical, not current quotations. Equity value / sales is not enterprise value / sales.

| Hypothetical share price | Illustrative equity value | Equity value / FY2025 sales |
| --- | ---: | ---: |
| SEK 10 | SEK 3.44 billion | 11.4x |
| SEK 25 | SEK 8.60 billion | 28.4x |
| SEK 50 | SEK 17.19 billion | 56.8x |

Formula: share price × 343,867,416 / 302,800,000. These examples show why a promising component story needs a demanding valuation examination. They do not identify fair value. The next model must connect current enterprise value to achievable revenue, operating margins, reinvestment, dilution and the time before shareholders receive cash. A comparison with a much larger semiconductor company is not a substitute for that bridge.

## Checkpoints that can change the view

| Checkpoint | Evidence that would strengthen the thesis | Evidence that would weaken it |
| --- | --- | --- |
| Q4 2026 LiDAR shipments | Disclosed deliveries and recognized product sales | Material shipment delays or reduced scope |
| 2027 customer ramps | Repeat production orders and improving product economics | Qualifications repeatedly pushed out |
| Manufacturing expansion | Milestones met with disciplined spending and usable output | Cost overruns or weak utilization |
| Cash conversion | Product growth accompanied by less cash consumption | Growing receivables or cash burn without corresponding shipments |
| Per-share value | Returns justify financing and incentive dilution | Ownership dilution outruns the improvement in business value |

The [financial calendar](https://www.sivers-semiconductors.com/investors/financial-calendar/) lists **26 November 2026** for Q3 results and **25 February 2027** for Q4 results. The [EGM notice](https://www.sivers-semiconductors.com/press/notice-to-attend-an-extraordinary-general-meeting-of-sivers-semiconductors-ab/) schedules a meeting on **22 October 2026**, including incentive-program proposals. Scheduled events are not completed outcomes.

## Source attribution and claim verification

| Source layer | Public source | How we use it |
| --- | --- | --- |
| Idea origin | [Serenity / @aleabitoreddit](https://x.com/aleabitoreddit) | Credit for the SIVE idea selected by the reader; a hypothesis to investigate |
| Thesis reconstruction | [Serenity Tracker SIVE deep dive](https://serenitytrades.com/deep-dive/sive), with its [methodology](https://serenitytrades.com/about) | Secondary summary of the bottleneck and commercialization argument; not verified holdings or verbatim posts |
| Dated secondary coverage | [Lookonchain, 28 August 2026](https://lookonchain.com/feeds/70304) | A traceable reference to commentary around the opportunity pipeline and production conversion |
| Primary operating evidence | Sivers reports, regulatory notices and announcements linked throughout this memo | Reported results and the issuer's own statements, with dates and forward-looking claims identified |
| Our analytical work | This report's calculations, debate and checkpoints | Independent interpretation of those sources, with missing evidence exposed |

We have not reconstructed Serenity's full original post history. Therefore **“most recommended,” exact mention counts, current conviction and current holdings remain unverified**. The tracker describes SIVE as a leading conviction idea; that characterization is attributed to the tracker, not an independently established ranking. An original-post audit would need a complete collection window and a definition separating recommendations, mentions, replies and duplicates.

We also checked [Nanalyze's August SIVE discussion](https://www.nanalyze.com/videos/is-sivers-the-next-big-semiconductor-stock/) as a counterpoint. Its conclusions are secondary commentary; we do not adopt unverified allegations or historical performance claims. The business case is tested against operating evidence rather than the popularity of either a bullish or bearish commentator.

## What is still open

- Independently verified current quote with exchange, currency and timestamp.
- Updated issued, treasury and fully diluted share counts after subsequent actions.
- A reconciled current cash/debt/capex bridge and valuation model.
- Counterparty confirmation of material programs where publicly available; company announcements are attributed to the issuer.
- Customer concentration, production margins, yields and firm order values where not disclosed in the reviewed material.
- A reproducible audit of the original Serenity claims.

Update 2026-10-22: the current quote, share count, cash position and a scenario valuation are now recorded in the dated valuation section below; customer concentration and the Serenity post audit remain open.

These gaps limit the investment conclusion. They do not prevent following a measurable business thesis. The next report update should compare actual shipments, financial results and financing with these checkpoints rather than merely repeating the AI-photonics narrative.

## Where the laser sits: one campus, four layers, and the money that reaches Sivers

Added: 2026-10-22. Content updated: 2026-10-22. Evidence reviewed: 2026-10-22. Review checkpoint: 2026-11-26. All dollar figures are **industry estimates (est.)**, not Sivers disclosures; operators are **speculated**.

On the web page this section opens with a responsive HTML diagram; the static export is available as a [cutaway poster (PNG)](https://theclaudetrader.com/stock-analysis/assets/sive-cutaway-20261022.png).

| Layer | What it is | Vendor | Unit cost (est.) |
| --- | --- | --- | --- |
| 1 · AI data center | One 1 GW campus, with chips | Operator: Meta or Amazon (speculated, via Jabil and Ayar Labs analyst mapping) | $35 to 50B |
| 2 · AI server rack | NVIDIA GB300 / Rubin-class NVL72 rack, 72 GPUs, ~130 kW | Assembled by Jabil, Foxconn or Dell | $3 to 4M |
| 3 · Optical transceiver | 1.6T OSFP linear-receive module; 3 to 4 per GPU across the scale-out fabric | Jabil LRO module with Sivers DFB lasers ([confirmed, 15 Apr 2026](https://www.sivers-semiconductors.com/press/sivers-semiconductors-collaborates-with-jabil-on-energy-efficient-1-6t-pluggable-optical-transceiver-module/)) | $1,500 to 2,000 |
| 4 · Laser die | InP DFB laser, 1310 nm, 2 to 4 per module | Sivers Photonics, Glasgow | $5 to 30 |

**Money flow for one campus (est.):** $40B operator spend → $18 to 22B to racks (6,000 to 7,500 racks × $3M, 1 GW facility at PUE 1.1 to 1.3) → about $3B to transceivers (~500k GPUs × 3 to 4 modules × $1,500 to 2,000, both link ends plus fat-tree tiers) → **$25 to 60M to Sivers lasers** (2 dies × ~$15 per module × 1.5 to 2M modules, if Sivers supplied every laser; point estimate ~$45M). Assumptions: ~130 kW per rack, 3 to 4 scale-out modules per GPU, no co-packaged optics. One campus fully won is therefore worth one to two years of Sivers' FY2025 net sales of about SEK 307M (~$32M).

Sources: [SemiAnalysis GB200 architecture](https://newsletter.semianalysis.com/p/gb200-hardware-architecture-and-component) (rack power, optics per GPU); [NBER working paper 35865](https://www.nber.org/papers/w35865) and [Bloomberg Law on the Meta Texas campus](https://news.bloomberglaw.com/tech-and-telecom-law/meta-blackrock-plan-to-invest-in-giant-texas-data-center) (campus cost with chips); [Ascent Optics 1.6T guide](https://ascentoptics.com/blog/osfp-1-6t-optical-transceiver-guide/) and [FS.com 1.6T listings](https://www.fs.com/c/1.6t-osfp-5597) (module prices); [Credo Oz8](https://credosemi.com/products/silicon-photonics/oz8/) and [OpenLight DR8 PIC](https://www.prnewswire.com/news-releases/openlight-announces-sample-availability-of-its-first-1-6tb-dr8-dfb-based-photonic-integrated-circuit-302411698.html) (lasers per module); [800G module BOM discussion](https://www.hytoptodevice.com/blog-detail/800g-optical-modules-where-is-the-market-heading-a-2026-deep-dive) (die price).

## Competitors and market share (22 October 2026)

Added: 2026-10-22. Content updated: 2026-10-22. Evidence reviewed: 2026-10-22. Review checkpoint: 2026-11-26. Revenue figures per competitor are **analyst estimates** for 2026 datacom InP lasers unless marked otherwise.

On the web page this section opens with a responsive HTML diagram; the static export is available as a [competitor map poster (PNG)](https://theclaudetrader.com/stock-analysis/assets/sive-competitors-20261022.png).

| Company | Role | Est. 2026 datacom laser revenue | Sells merchant CW DFB dies? |
| --- | --- | ---: | --- |
| Lumentum | EML leader; CW lasers for its own transceivers from fiscal Q3 2026 | ~$1.2B | EML yes; CW mostly captive so far |
| Coherent | Merchant; 400 mW CW-DFB for silicon photonics and CPO | ~$600M | Yes |
| Broadcom | Largely captive; CW-DFB capacity leader | ~$500M | Partly |
| Sumitomo Electric | Merchant; CW-DFB capacity co-leader | ~$400M | Yes |
| Mitsubishi Electric | Merchant EML | ~$400M | EML mainly |
| LandMark / LuxNet | Merchant CW-DFB foundry (Taiwan) | ~$150M | Yes |
| HG Genuine, Yuanjie | Merchant (China) | ~$150M | Yes |
| MACOM | Merchant | ~$100M | Yes |
| Nokia (Infinera InP) | Captive | ~$100M | No |
| Applied Optoelectronics | Captive | ~$80M | No |
| Source Photonics | Captive | ~$60M | No |
| OpenLight / Tower | DFB grown into the silicon-photonics wafer | ~$30M | Platform, not dies |
| **Sivers Photonics** | Merchant | **~$10M** (SEK 93.4M, all photonics, 2025, [annual report](https://www.sivers-semiconductors.com/wp-content/uploads/2026/05/Sivers_annualreport_2025_final.pdf)) | Yes |

On the web page this section opens with a responsive HTML diagram; the static export is available as a [market-share poster (PNG)](https://theclaudetrader.com/stock-analysis/assets/sive-market-share-20261022.png).

[TrendForce (June 2026)](https://www.semiconductor-today.com/news_items/2026/jun/trendforce-040626.shtml) puts Broadcom, Lumentum and Sumitomo at about 55% of EML plus CW-DFB supply, and Broadcom, Sumitomo, Coherent and LandMark at about 74% of CW-DFB capacity, with Lumentum's EML capacity booked through 2027. TrendForce reports capacity (50.7M dies a month) rather than dollars. At $8 to 10 per die that is roughly $3B a year, which would make Sivers' share about **0.3%**, lower still for datacom alone.

**The market size is the weakest input.** The only published dollar series is [Future Market Insights](https://www.futuremarketinsights.com/reports/cw-dfb-laser-chip-market): $579M for CW-DFB chips (2025, all uses) growing to about $1B by 2030. The $3B figure is unit-derived from TrendForce. The scenarios below therefore show the share Sivers would need under both denominators.

## Scenarios: 2030 photonics revenue under bust, base and AI boom (22 October 2026)

Added: 2026-10-22. Content updated: 2026-10-22. Evidence reviewed: 2026-10-22. Review checkpoint: 2026-11-26. **This analysis, not company guidance.** The published consensus is a single house ([Redeye via MarketScreener](https://www.marketscreener.com/quote/stock/SIVERS-SEMICONDUCTORS-AB-61747388/finances/)): group sales SEK 273M (2026), 458M (2027), 621M (2028); [Fintel](https://fintel.io/sfo/se/sive) shows 689M for 2028. Consensus implies photonics of roughly $30M in 2028.

On the web page this section opens with a responsive HTML diagram; the static export is available as a [scenarios poster (PNG)](https://theclaudetrader.com/stock-analysis/assets/sive-scenarios-20261022.png).

| Scenario | Arithmetic | 2030 photonics revenue | Share needed at $1B TAM / $3B TAM | Versus consensus |
| --- | --- | ---: | --- | --- |
| Bust | $1.5B TAM × 0.5% = 1M dies × $7.5 = $7.5M datacom + $10M LiDAR and sensing | ~$18M | 0.8% / 0.3% | Roughly the consensus path |
| Base | $2.5B TAM × 3% = 9M dies × $8 = $75M + $15M | ~$90M | 7.5% / 2.5% | Bullish: needs ~70% a year after 2028 |
| AI boom | $4B TAM (CPO external light sources add $1B+) × 8% = 40M dies × $8 = $320M + $20M | ~$340M | 32% / 8% | About 5× consensus |

Implied growth from the SEK 93.4M (~$9.8M) 2025 base over five years: bust 1.8× (~12% a year), base 9× (~55% a year), boom 34× (~100% a year). The photonics run-rate in H1 2026 was SEK 33.7M, about SEK 67M annualised, so every scenario starts from a lower point than 2025.

What decides the case:

- **Share, not market size.** Going from 0.3% to 3% is the whole base case against suppliers that hold most CW-DFB capacity.
- **CPO versus pluggables.** In a pluggable the Sivers die is a $15 commodity. Analyst estimates put ultra-high-power DFB lasers at $50 to 100 each with about eight per external light source, or $300 to 600 of laser content per GPU under co-packaged optics versus $30 to 60 in pluggables ([Mizuho via secondary](https://x.com/TheValueist/status/1993439612551758073), [siliconmatter](https://siliconmatter.substack.com/p/sivers-semiconductors-a-130m-mcap)). No Ayar Labs SuperNova price is public. Ayar's volume qualification is targeted for H2 2027, so none of it appears before 2028. The boom case is a CPO case.
- **Capacity ceiling.** The [Q2 2026 CEO letter](https://www.sivers-semiconductors.com/2026/08/27/vickram-vathulyas-letter-to-shareholders-interim-report-q2-2026/) says the expanded Glasgow facility is functional by end-2027 and targets a 1:2 ratio of internal to partner-foundry capacity. Revenue above a few tens of millions of dollars in 2027 depends on the partner line.
- **LiDAR.** The [20 October 2026 release](https://www.sivers-semiconductors.com/press/strategic-lidar-customer-initiates-production-orders-with-sivers-semiconductors/) confirms production orders: "Sivers will supply the continuous-wave distributed feedback (CW DFB) lasers and optical amplifiers", shipments from Q4 2026, production in 2027, "estimated cumulative revenue potential of USD 53 million to USD 138 million over the customer's product lifecycle" (programme first announced March 2026). This supersedes the USD 28 to 53 million cumulative 2026 to 2030 figure in the Q4 2025 report used in earlier drafts. The scenarios' LiDAR-and-sensing lines ($10 to 20M a year) sit inside that lifecycle range; the first order value is not disclosed.
- **Pipeline.** The $1.2B opportunity pipeline (July 2026) is non-binding, covers 2026 to 2030 and both divisions, and was $453M at end-2025 ([Q4 2025 report](https://www.sivers-semiconductors.com/press/sivers-semiconductors-ab-publ-publishes-interim-report-q4-october-december-2025/)).

Additional sources: [LightCounting, May 2026](https://www.lightcounting.com/newsletter/en/may-2026-silicon-photonics-lpolro-and-npocpo-377); [Photoncap on the light-source market](https://photoncap.net/p/the-silicon-photonics-light-source); [Lumentum Q2 FY26](https://www.semiconductor-today.com/news_items/2026/feb/lumentum-090226.shtml); [SEQH Research](https://www.seqhresearch.com/p/sivers-semiconductors-sive-inp-laser); [Ayar Labs and Jabil SuperNova sled](https://ayarlabs.com/jabil-server-sled-with-64-ayar-labs-supernova-light-sources/).

## Valuation: market data and scenario values as of 22 October 2026

Added: 2026-10-22. Content updated: 2026-10-22. Evidence reviewed: 2026-10-22. Review checkpoint: 2026-11-26. This resolves the current-quote, share-count and cash items listed as open on 20 October, with the limits stated below. FX est. at EUR 11.0 and USD 9.5 SEK.

On the web page this section opens with a responsive HTML diagram; the static export is available as a [valuation poster (PNG)](https://theclaudetrader.com/stock-analysis/assets/sive-valuation-20261022.png).

| Item | Value | Source |
| --- | --- | --- |
| Share price | €2.06 ≈ SEK ~23 on 22 Oct 2026, 12:31 CET; 52-week high €10.2 (June) | [boerse.de](https://www.boerse.de/aktien/Sivers-Semiconductors-Aktie/SE0003917798) |
| Shares outstanding | ~355.1M after the July actions (319.95M at 30 Jun + 12.28M July issue + Bootstrap warrant exercise); EGM notice: 356.7M issued, 12.9M in treasury | [Q2 2026 report](https://www.sivers-semiconductors.com/wp-content/uploads/2026/08/Sivers-Interim-report-Q226.pdf), [boerse-express](https://www.boerse-express.com/news/articles/sivers-semiconductors-aktie-1215-prozent-einbruch-auf-242-euro-935128) |
| Market cap | SEK ~8.1B ≈ $0.85B | derived |
| Net cash | SEK ~0.7B est.: 62.7M at 30 Jun + SEK 700M raise at 57 (July); USD 5M term loan repaid August; convertible converted; no interest-bearing debt | Q2 report, [700 MSEK issue](https://www.sivers-semiconductors.com/press/sivers-semiconductors-has-resolved-on-a-directed-share-issue-of-shares-amounting-to-approximately-sek-700-million/) |
| Enterprise value | SEK ~7.3B ≈ $0.77B | derived |
| EV / sales | 24× on 2025 (SEK 306.6M restated), 27× 2026e (273M), 16× 2027e (458M), 12× 2028e (621M) | single-analyst consensus via MarketScreener |
| Consensus EBITDA | 2026e −173M, 2027e +33M, 2028e +131M | [Redeye Q2 review](https://www.redeye.se/research/1181326/sivers-semiconductors-q2-review-a-deliberate-trade-off-ahead-of-the-2027-ramp) |
| Redeye DCF | base SEK 7.4, range 2.1 to 14.8, written at SEK 27 to 30 | same |
| Segments | Photonics 2025 SEK 93.4M (2024: 80.1M); Wireless 213.1M; H1 2026 Photonics 33.7M (−23%), Wireless 81.9M (−19%) | annual report 2025, Q2 2026 report |
| Dilution | Sep 2025 SEK 95M; Apr 2026 8.62M shares at SEK 14.5 (125M); Jul 2026 12.28M at SEK 57 (700M, ~3.3% fully diluted); 22.8M shares to Bootstrap on conversion; share count up ~32% in 20 months | [April issue](https://www.sivers-semiconductors.com/press/sivers-semiconductors-har-beslutat-om-en-riktad-nyemission-av-aktier-om-cirka-125-msek/), Q1 2026 report |
| Peers, EV/sales trailing → forward | Lumentum 25× → ~12× (EV $76B); Coherent 6.8× → ~4.6× (EV $48B); Aeva 31× → ~14× (EV $0.66B); POET pre-revenue (cap $0.98B, EV $0.19B) | [stockanalysis.com](https://stockanalysis.com/stocks/lite/statistics/), 22 Oct 2026 |

**Scenario values.** 2030 group revenue = photonics scenario + ~$45M wireless (SEK ~430M, extrapolated from consensus 2028); EV at 6× and 12× sales; plus $0.08B cash; divided by 355M shares; discounted four years at 12% (×0.64). No further dilution assumed.

| Scenario | 2030 group revenue | EV at 6× / 12× | SEK per share in 2030 | SEK per share today |
| --- | ---: | --- | --- | --- |
| Bust | $63M | $0.38B / $0.76B | 12 / 23 | 8 / 14 |
| Base | $135M | $0.81B / $1.62B | 24 / 46 | 15 / 29 |
| AI boom | $385M | $2.3B / $4.6B | 64 / 126 | 41 / 80 |

Reading: at SEK ~23 the market already prices the base case at a Lumentum-like multiple. The consensus path alone does not support the price. The boom case offers two to four times, but needs CPO light sources at volume from 2028 and the partner foundry to deliver, and would likely need a larger raise or the photonics listing. Still open after 22 October: the first LiDAR order value and 2027 run-rate (lifecycle USD 53 to 138M is not a one-year forecast), customer concentration, Q3 cash, and a second analyst house. Carnegie, Pareto and Modular Finance estimates were not obtainable; "consensus" here is one house. The byNordic SPAC letter of intent (Aug 2024, paused Nov 2024) never disclosed a valuation.

## Broker and ratings data, IBKR Mobile, captured 2026-10-22

Added: 2026-10-22. Content updated: 2026-10-22. Evidence reviewed: 2026-10-22. Review checkpoint: 2026-11-26. Captured from IBKR Mobile on a read-only login between 12:48 and 12:55 Europe/Luxembourg (+02:00). Every row carries the as-of date IBKR showed and the capture timestamp of the screenshot it was read from. Third-party data (Morningstar, LSEG, TipRanks) via IBKR, not guaranteed by IBKR. Morningstar fields carry the "Q" marker: quantitative model output, not an analyst-written report.

![IBKR Mobile Morningstar quantitative summary for Sivers, captured 2026-10-22 12:49:54 CEST](https://theclaudetrader.com/stock-analysis/assets/sive-ibkr-morningstar-20261022-124954.png)

| Item | Value | IBKR as-of | Captured (Luxembourg) |
| --- | --- | --- | --- |
| Morningstar quantitative rating | 2 stars | 2026-10-22 | 2026-10-22T12:49:54+02:00 |
| Economic moat (Q) | None | 2026-10-21 (text date) | 2026-10-22T12:49:54+02:00 |
| Fair value estimate (Q) | SEK 16.23; stock at a 49% premium | FVE 2026-10-19; text 2026-10-21 | 2026-10-22T12:50:23+02:00 |
| Uncertainty / financial health (Q) | Very High / Moderate | 2026-10-22 | 2026-10-22T12:49:54+02:00 |
| Earnings yield / book value yield | −3.1% (bottom 20% of global peers) / 12.5% (bottom 10%) | 2026-10-21 | 2026-10-22T12:50:31+02:00 |
| Momentum note (verbatim) | "This share class has outperformed the broader universe over the past year. This outperformance may signify a bull trap, in light of other detractors from our model." | 2026-10-21 | 2026-10-22T12:50:31+02:00 |
| LSEG consensus | Sell, based on 1 analyst (1 Sell) | 2026-10-22 | 2026-10-22T12:51:04+02:00 |
| LSEG price target | mean SEK 7.70 (−67.09% vs price); high 8.00 (−65.81%), median 7.70, low 7.40 (−68.38%); implied price ≈ SEK 23.4 | 2026-10-22 | 2026-10-22T12:51:04+02:00 |
| TipRanks consensus and target | Not enough data | 2026-10-22 | 2026-10-22T12:51:18+02:00 |
| TipRanks individual rating | Tim Savageaux, Northland Securities, Buy, target $1 (ADR basis), "9 months ago" (≈ Jan 2026) | 2026-10-22 | 2026-10-22T12:51:18+02:00 |
| TipRanks news sentiment | Not enough data | 2026-10-22 | 2026-10-22T12:51:44+02:00 |
| ESG (Company tab, provider not named) | E 1 · S 1 · G 6; overall 2; controversies Low (score 10) | 2026-10-22 | 2026-10-22T12:49:00+02:00 |
| Ownership | Institutions 17.54%; strategic entities 21.34%; others 61.12% | 2026-10-22 | 2026-10-22T12:48:54+02:00 |
| Quote header (SFB) | EPS −1.01; beta 1.594; market cap "8B" SEK; IV 101.4%; put/call interest 0.59. Price, volume and 52-week range blank (no Stockholm market-data subscription) | 2026-10-22 | 2026-10-22T12:48:18+02:00 |
| Key ratios (USD, TTM) | P/S 28.6; P/B 7.51; P/E normalized −31.82; ROE −29.86%; ROA −21.09%; current ratio 1.64; book value per share 0.30; cash per share 0.02 | interim 2026-06-30; annual 2025-12-31 | 2026-10-22T12:53:42+02:00 to 12:53:49+02:00 |
| Income statement, quarter to 2026-06-30 (USD m) | Revenue 5.13 (−12.4% y/y); gross profit −1.87; operating income −11.14; net income −10.96; diluted EPS −0.04 | Jun 30, 2026 | 2026-10-22T12:53:30+02:00 to 12:53:36+02:00 |
| LSEG annual forecasts (SEK, read from bar charts, approximate) | Revenue 2026E ~270M, 2027E ~460M, 2028E ~620M; EBITD 2026E ~−165M, 2027E ~+25M, 2028E ~+130M; EPS 2026E ~−0.49, 2027E ~−0.08, 2028E ~+0.07; 2025 shown pre-restatement at 304.1M | chart as-of 2026-10-22 | 2026-10-22T12:52:30+02:00 to 12:52:40+02:00 |
| LSEG quarterly forecasts | Q3 2026E revenue ~57M, Q4 2026E ~98M; Q2 2026 actual 53.8M vs ~60M estimate | 2026-10-22 | 2026-10-22T12:52:01+02:00 to 12:52:11+02:00 |
| News tab | 2026-10-20 Reuters / Dow Jones / PR Newswire: "Strategic LiDAR Customer Initiates Production Orders With Sivers Semiconductors" | 2026-10-20 | 2026-10-22T12:54:22+02:00 |
| SSMIY unsponsored ADR (PINK) | Last 7.30 USD (frozen, pre-US-open); bid/ask 7.11/8.10, spread 12.2%; 52-week 10.21/6.69; implies about 3 ordinary shares per ADR (inference) | 2026-10-22 | 2026-10-22T12:55:22+02:00 |

![IBKR Mobile LSEG analyst consensus and price targets for Sivers, captured 2026-10-22 12:51:04 CEST](https://theclaudetrader.com/stock-analysis/assets/sive-ibkr-lseg-20261022-125104.png)

![IBKR Mobile TipRanks panel for Sivers, captured 2026-10-22 12:51:18 CEST](https://theclaudetrader.com/stock-analysis/assets/sive-ibkr-tipranks-20261022-125118.png)

![IBKR Mobile ESG and ownership panel for Sivers, captured 2026-10-22 12:49:00 CEST](https://theclaudetrader.com/stock-analysis/assets/sive-ibkr-esg-20261022-124900.png)

How this squares with the rest of the page: the LSEG single analyst is consistent with the Redeye figures used in the valuation section (SEK 7.4 fair value, revenue 273/458/621M), so "consensus" on this page is one house seen through two data vendors. Morningstar's SEK 16.23 is a quantitative model value with very high uncertainty, not a research call. The one TipRanks Buy is the January 2026 Northland record already listed under analyst records.

**Not captured:** no live Stockholm quote (price, volume, 52-week range) on the read-only login; no short-interest or securities-lending data for this listing; no events or earnings-calendar tab in the Company layout; the Morningstar "Full Report" PDF was not downloaded; the 2DG (Xetra) line was not opened. The 52 capture screenshots are listed in the [image manifest](https://theclaudetrader.com/stock-analysis/SIVE.assets.json); four are embedded above.

## Leadership and governance: the CEO, his letters, and the NRE story (22 October 2026)

Added: 2026-10-22. Content updated: 2026-10-22. Evidence reviewed: 2026-10-22. Principal source published: 2026-08-27. Review checkpoint: 2026-11-26.

**Who runs the company.** Vickram Vathulya is not a founder; Sivers dates to 1951. He joined from outside as President and CEO on 19 August 2024 after a search by Hunt Scanlon ([appointment](https://www.sivers-semiconductors.com/press/sivers-semiconductors-appoints-vickram-vathulya-as-new-president-ceo/)). He replaced Anders Storm, CEO since 2016, who [resigned on 26 March 2024](https://news.cision.com/sivers-semiconductors/r/sivers-semiconductors-announces-resignation-of-ceo-anders-storm,c3951931) "for new challenges outside the Company" with no further reason. Chair: Bami Bastani. After the [15 June 2026 AGM](https://www.sivers-semiconductors.com/press/bulletin-from-sivers-semiconductors-ab-publs-annual-general-meeting-on-15-june-2026/) the board is Bastani, Todd Thomson, Karin Raj, Joakim Nideborn and Helena Svancar; Erik Fällström (largest holder, 13.86% at end-2025), Tomas Duffy and Keith Halsey left. CFO Heine Thorsgaard since September 2025.

**Track record.** Philips/NXP (1999 to 2008), then twelve years at Maxim Integrated ending as VP and GM of Core Products, then President of Nuvotronics, a millimetre-wave subsidiary of Cubic, from November 2022 to August 2024. A divisional operator at a large analog house; no prior public-company CEO role, no photonics background, no controversies found. No independently verifiable outcome from Nuvotronics.

**Promises versus delivery.**

| Letter or release | Promise | What happened |
| --- | --- | --- |
| [Dec 2024 Ayar Labs release](https://www.sivers-semiconductors.com/press/sivers-semiconductors-and-ayar-labs-to-expand-their-partnership-on-enabling-high-volume-manufacturing-of-optical-i-o-solutions-for-scalable-cost-effective-ai-infrastructure/) | Shipment volumes "expected to ramp in the next 18 to 24 months" | Ayar is absent from the Q4 2025, Q1 2026 and Q2 2026 letters; no volume order announced; photonics hardware −23% in H1 2026 |
| Q4 2025 (26 Feb 2026) | LiDAR ramp Q4 2026; pipeline $453M; breakeven at $50 to 55M revenue "within two years" | **Delivered:** [production orders announced 20 Oct 2026](https://www.sivers-semiconductors.com/press/strategic-lidar-customer-initiates-production-orders-with-sivers-semiconductors/), two days before the Q4 window; "shipments will begin in Q4 2026 against rolling production orders". Breakeven still implies ~SEK 500M revenue versus SEK 116M in H1 2026 |
| [Q1 2026 letter (29 May 2026)](https://www.sivers-semiconductors.com/2026/05/29/vickram-vathulyas-letter-to-shareholders-interim-report-q1-2026/) | "On track to our full-year revenue growth plan"; ALL.SPACE orders imminent | ALL.SPACE $8.2M order came; H1 sales −20%, Q2 −12%, and the growth plan was dropped one quarter later |
| [Q2 2026 letter (27 Aug 2026)](https://www.sivers-semiconductors.com/2026/08/27/vickram-vathulyas-letter-to-shareholders-interim-report-q2-2026/) | LiDAR orders "expected imminently" for Q4 2026; revenue inflection Q4 2026; Jabil beta Q4 2026, production orders H1 2027; telco FWA by end-2026; US listing prep done H1 2027; Glasgow functional end-2027 | LiDAR orders **delivered 20 Oct 2026** (first milestone of this letter met). The rest is testable at the Q3 report on 26 Nov 2026 and the Q4 report on 25 Feb 2027 |
| [byNordic SPAC LOI (6 Aug 2024)](https://www.sivers-semiconductors.com/press/sivers-semiconductors-enters-into-letter-of-intent-with-bynordic-acquisition-corporation-for-business-combination-of-sivers-photonics-subsidiary/) | List Sivers Photonics via SPAC | [Paused Nov 2024](https://www.marketscreener.com/quote/stock/SIVERS-SEMICONDUCTORS-AB-61747388/news/Sivers-Semiconductors-pauses-discussions-to-merge-subsidiary-with-spac-company-48318417/); replaced by a potential US dual listing (15 Apr 2026); not delivered |

Pattern: hard milestones move one or two quarters right each letter while the non-binding pipeline figure ($453M → $799M → $1.2B) is the headline.

**The NRE question, with the numbers (SEK m, restated).**

| Period | Hardware | NRE | Total | Photonics HW | Wireless HW |
| --- | ---: | ---: | ---: | ---: | ---: |
| FY2024 | 131.3 | 87.3 | 219.2 | 80.1 | 51.2 |
| FY2025 | 127.2 | 178.7 | 306.6 | 93.4 | 33.8 |
| Q2 2025 | 25.7 | 35.5 | 61.4 | 17.2 | 8.6 |
| Q2 2026 | 29.2 | 24.6 | 53.8 | 15.9 | 13.3 |
| H1 2026 vs H1 2025 | 56.8 vs 66.9 (−15%) | 58.8 vs 77.1 (−24%) | 115.6 vs 144.3 (−20%) | 33.7 vs 43.9 (−23%) | 23.1 vs 23.0 |

All NRE is Wireless; Photonics carries no NRE after the restatement, so "shifting from NRE to product" is a Wireless story and says nothing about AI-optics demand. 2025 growth was NRE doubling on US defence programmes (three Wireless customers were 60% of revenue) while hardware fell. Contract liabilities (customer prepayments for new development work) fell from SEK 32.3M to 11.6M during 2025 while contract assets (NRE recognised but unbilled) rose from 54.7M to 100.4M: fewer new prepaid contracts and more unbilled work in progress, which is what a development pipeline running off looks like. The Q2 "product revenue +18%" headline is Wireless hardware; photonics hardware fell 8% in Q2 and 23% in H1. Sources: annual report 2025 Note 6, Q2 2026 report p. 14, Q1 2026 report p. 13.

**Cash and dilution.** Cash SEK 43.5M at end-2025 (operating cash flow −57M), 26.6M at end-Q1 2026, 62.7M at end-Q2 before the SEK 700M raise closed (Q2 operating cash flow −70M). Headcount 130 (June 2026).

**Insider alignment.** Vathulya holds about 4.54M shares (~1.3%) plus 3.7M options, committed at hiring to buy 2M shares, and bought 70,000 more after the lock-up on 21 July 2026 at about SEK 38 to 42: a net buyer above today's price. Chair Bastani sold 275,000 shares the day the lock-up expired (16 July 2026) and director Thomson's Headwaters sold 950,000 by 22 July ([lock-up release](https://www.sivers-semiconductors.com/press/update-on-expiry-of-lock-up-undertaking-and-insider-transactions-in-sivers-semiconductors/)). A circulated report that the CEO sold 1.23M shares was false ([Boerse Express](https://www.boerse-express.com/news/articles/sivers-semiconductors-aktie-ceo-vickram-vathulya-kauft-nach-929034)). 2025 pay SEK 5.1M, no bonus.

**Red flags.**

- [May 2026 restatement](https://www.sivers-semiconductors.com/press/sivers-semiconductors-publishes-the-2025-annual-report-adjusting-prior-reported-financials-as-preparation-for-a-potential-dual-listing-in-the-united-states/): 2024 net sales cut from 243.7 to 219.2M (−10%), 2024 net loss widened from −116.3 to −183.9M, 2025 EBIT from −141.3 to −177.8M, equity cut by SEK 127M. Causes: NRE recognised too early, hardware timing, distributor returns, inventory write-ups, a loss-making NRE contract, understated amortisation and a SEK 28.4M impairment of capitalised Wireless development. Framed as a PCAOB audit uplift; these are IFRS errors in the core revenue model, under the current CEO for 2025.
- Share count up ~32% in 20 months; the [22 October 2026 EGM](https://www.sivers-semiconductors.com/press/notice-to-attend-an-extraordinary-general-meeting-of-sivers-semiconductors-ab/) votes on a new option plan and Series C shares.
- [Dagens Industri, 31 Aug 2026](https://www.boerse-express.com/news/articles/sivers-semiconductors-aktie-dagens-industri-ruft-zum-verkauf-auf-943310) called the Q2 report catastrophic and said sell; the stock fell 15% that day and 27% over seven sessions.
- Share price under his tenure: low SEK 1.76 (13 Nov 2024), high SEK 110 (3 Jun 2026), about SEK 23 to 24 now ([TradingView](https://www.tradingview.com/symbols/OMXSTO-SIVE/)).

**Balanced verdict.** For: real Wireless wins (ALL.SPACE $8.2M, Tachyon $3M, CHIPS Act programmes), genuine photonics partner names (Jabil, GlobalFoundries, POET, O-Net), a SEK 700M raise executed near the top, no debt, and the CEO buying with his own money while the chair and a director sold. Against: the group has not grown once restated; the AI division shrank 23% in H1 2026; the Ayar ramp promise vanished; the SPAC never closed; the restatement found NRE revenue overstated; new development bookings are drying up; guidance slides right every quarter. He reads as a credible divisional operator with aligned incentives, not a proven public-company CEO, and the letters run ahead of the numbers. The LiDAR production orders of 20 October 2026 are the first delivered milestone of the Q2 letter and count in his favour; the first hard numerical test is the Q3 report on 26 November 2026 and whether LiDAR shipments and Jabil orders show up in Q4.

## Fact audit: corrections made on 22 October 2026

Added: 2026-10-22. An adversarial check of the diagram and scenario inputs found these errors in earlier drafts; the published figures above are the corrected ones.

| Claim in earlier draft | Correction | Basis |
| --- | --- | --- |
| 8 laser dies per 1.6T module | 2 to 4 CW lasers per DR8 photonic IC (8 optical lanes, not 8 lasers) | Credo Oz8, OpenLight DR8 |
| 1 to 2 optical modules per GPU | 3 to 4 per GPU across the fabric (both link ends, fat-tree tiers) | SemiAnalysis |
| 1 GW campus $10 to 15B | $35 to 50B with chips; $14B is development only | NBER, Bernstein, Bloomberg |
| 1.6T module $1,000 to 1,500 | $1,500 to 2,000 (list $1,300 to 2,400) | Ascent Optics, FS.com |
| Sivers die $2 to 10 | $5 to 30 est.; no public price | vendor BOM examples |
| GB200 rack with 1.6T modules | GB200 uses 400G/800G; 1.6T belongs to GB300/Rubin, so the rack is labelled GB300/Rubin-class | SemiAnalysis |
| DR8 70 to 100 mW tie to the Jabil module | Release names only "DFB lasers"; the specific part is inference and was removed | Sivers release 15 Apr 2026 |
| CW-DFB datacom TAM ~$3B | No published dollar source; shown as $1B (published) versus $3B (unit-derived) | FMI, TrendForce |
| Base case versus consensus | Labelled bullish versus the single-analyst path, not a mid case | MarketScreener, Fintel |
| LiDAR cumulative figure USD 28 to 53M (Q4 2025 report) | Updated to USD 53 to 138M over the product lifecycle per the 20 Oct 2026 release; earlier figure superseded | Sivers release 20 Oct 2026 |

Diagrams on the web page are responsive HTML built from the tables in this report; the PNG posters linked under each are static exports of the working design canvases and may clip a footer line.

## Method and update log

Company disclosures establish reported figures and what management said. Our interpretations and hypothetical calculations are labeled separately. The local evidence manifest records retrieval status and hashes; a web-readable source may still have a failed local download, which is disclosed rather than silently treated as archived. The maintained narrative is this dated report, not an automatically generated stock rating.

- **2026-10-20:** Initial operating-evidence review; included today's LiDAR production-order release. Valuation inputs and original social-post audit remain open.
- **2026-10-20, attribution revision:** Disclosed Serenity and the tracker prominently, added direct source links and a detailed investment-underwriting framework. Distinguished original commentary, secondary reconstruction, company evidence and our own analysis.
- **2026-10-22, third pass:** Replaced the five embedded diagram PNGs with responsive HTML diagrams (nested cutaway, competitor bar table, share and TAM bars, scenario cards, valuation key-value panel) that reflow on phones; PNG posters remain as downloads. Data unchanged.
- **2026-10-22, second pass:** Added the IBKR Mobile broker and ratings capture (Morningstar quantitative, LSEG, TipRanks, ESG, ownership, key ratios, LSEG forecasts) with as-of and capture timestamps, and worked the 20 October LiDAR production-order release into the promises table, scenarios, valuation open items, conclusion and fact audit.
- **2026-10-22:** Added the campus money-flow cutaway, competitor and market-share review, 2030 scenarios against consensus, a dated valuation with scenario values per share, the leadership and NRE analysis, and a fact-audit table of corrections. Diagram exports added as dated assets.

## Update: what drove Q2 hardware growth

Added: 2026-10-21. Content updated: 2026-10-21. Evidence reviewed: 2026-10-21. Principal source published: 2026-08-27. Financial period ended: 2026-06-30. Review checkpoint: 2026-11-26.

Wireless hardware sales rose from SEK 8.6m to 13.3m (about 55%), while Photonics hardware fell from SEK 17.2m to 15.9m. Total hardware grew from SEK 25.7m to 29.2m. Figures are rounded; prior-year comparisons are restated.

Management highlighted production builds for Tachyon Networks against an initial USD 3m fixed-wireless-access production order. It did not disclose each customer's contribution to Q2 revenue, so attributing a precise share of growth to Tachyon is not supported. ALL.SPACE's USD 8.2m order supports a 2027 ramp; the August update anticipated Jabil-related production in 2027. These future orders do not establish the source of Q2 sales already recorded.

NRE revenue fell from SEK 35.5m to 24.6m, more than offsetting hardware growth. Management cited resource reprioritisation toward future production, US defence budget delays, customer timing and currency effects. First-half hardware revenue fell from SEK 66.9m to 56.8m: the quarterly improvement is not yet an established growth trend.

Sources: [Q2 report, page 15](https://www.sivers-semiconductors.com/wp-content/uploads/2026/08/Sivers-Interim-report-Q226.pdf#page=15); [CEO's 27 August update](https://www.sivers-semiconductors.com/2026/08/27/vickram-vathulyas-letter-to-shareholders-interim-report-q2-2026/).

## Dated external analyst records

Added: 2026-10-20. Content updated: 2026-10-20. Evidence reviewed: 2026-10-20. Review checkpoint: 2026-11-20. These records are not a current blended consensus.

- Redeye, 2026-09-03: SEK 7.40 fair-value estimate maintained, verified through Finwire's secondary reporting. The original note was blocked (403). A fair-value estimate is not a Buy/Hold/Sell rating.
- Northland Securities, 2026-01-07: historical Buy / SEK 8 target, attributed to a TipRanks record and The Fly archive. Original research not obtained; not represented as a fresh recommendation.
- DNB Carnegie Access, 2026-09-01: coverage discontinued. No active target inferred.

Source links and retrieval limits: [analyst records](https://theclaudetrader.com/stock-analysis/SIVE.analysts.json).

## Information dates and future updates

This research has one canonical reader page: [Sivers Semiconductors](https://theclaudetrader.com/stock-analysis/SIVE). The old workspace address redirects there. This Markdown preserves the detailed research text. Section-level history is published as [SIVE.updates.json](https://theclaudetrader.com/stock-analysis/SIVE.updates.json), including added_at, content_updated_at, last_reviewed_at, source_published_at, source_period_end and review_after. Dates have day precision. Layout changes do not reset evidence dates. A review checkpoint indicates work due, not proof that newer information exists or that a forecast came true.


## Reader-provided visual summary — added 21 October 2026

[View the Q2 2026 infographic](https://theclaudetrader.com/stock-analysis/SIVE#infographic). Download timestamp: **21 October 2026 at 19:40:05 CEST (Luxembourg)**, from the original local file creation time. The image carries a Gemini Notebook watermark; its source publication time is unverified. The chart shows hardware components rather than total segment revenue, and the exact share of Q2 growth attributable to Tachyon was not disclosed. Refer to the sourced financial explanation on the page.


## Reader-provided video overview — added 21 October 2026

[Watch the video near the top of the analysis](https://theclaudetrader.com/stock-analysis/SIVE#video). Duration: 7 minutes 29 seconds. Download timestamp: **21 October 2026 at 19:55:44 CEST (Luxembourg)**, based on original local file creation time. Gemini Notebook watermark visible; source publication time unverified. The video is supplementary commentary and its claims have not been independently audited; use the dated issuer evidence in the written analysis.