Slides
Page 1
February 2026 Q4 Investor Presentation Nasdaq:INOD
Page 2
2 DisclaimersForward-Looking StatementsThis presentation contains forward-looking statements within the meaning of Section 21E of the Securities Exchange Act of 1934, as amended, and Section 27A of the Securities Act of 1933, as amended. These forward-looking statements include, without limitation, statements concerning our operations, economic performance, financial condition, developmental program expansion and position in the generative AI services market. Words such as “project,” “believe,” “expect,” “can,” “continue,” “could,” “intend,” “may,” “should,” “will,” “anticipate,” “indicate,” “predict,” “likely,” “estimate,” “plan,” “potential,” “possible,” “promises,” or the negatives thereof, and other similar expressions generally identify forward-looking statements.. These forward-looking statements are based on management’s current expectations, assumptions and estimates and are subject to a number of risks and uncertainties, including, without limitation, impacts resulting from ongoing geopolitical conflicts; investments in large language models; that contracts may be terminated by customers; projected or committed volumes of work may not materialize; pipeline opportunities and customer discussions which may not materialize into work or expected volumes of work; the likelihood of continued development of the markets, particularly new and emerging markets, that our services support; the ability and willingness of our customers and prospective customers to execute business plans that give rise to requirements for our services; continuing reliance on project-based work in the Digital Data Solutions (“DDS”) segment and the primarily at-will nature of such contracts and the ability of these customers to reduce, delay or cancel projects; potential inability to replace projects that are completed, canceled or reduced; our DDS segment’s revenue concentration in a limited number of customers; our dependency on content providers in our Agility segment; our ability to achieve revenue and growth targets; difficulty in integrating and deriving synergies from acquisitions, joint ventures and strategic investments; potential undiscovered liabilities of companies and businesses that we may acquire; potential impairment of the carrying value of goodwill and other acquired intangible assets of companies and businesses that we acquire; a continued downturn in or depressed market conditions; changes in external market factors; the potential effects of U.S. global trade and monetary policy, including the interest rate policies of the Federal Reserve; changes in our business or growth strategy; the emergence of new, or growth in existing competitors; various other competitive and technological factors; our use of and reliance on information technology systems, including potential security breaches, cyber-attacks, privacy breaches or data breaches that result in the unauthorized disclosure of consumer, customer, employee or company information, or service interruptions; and other risks and uncertainties indicated from time to time in our filings with the Securities and Exchange Commission (“SEC”). Our actual results could differ materially from the results referred to in any forward-looking statements. Factors that could cause or contribute to such differences include, but are not limited to, the risks discussed in Part I, Item 1A. “Risk Factors,” Part II, Item 7. “Management’s Discussion and Analysis of Financial Condition and Results of Operations,” and other parts of our Annual Report on Form 10-K, filed with the SEC on February 26, 2026, and in our other filings that we may make with the SEC. In light of these risks and uncertainties, there can be no assurance that the results referred to in any forward-looking statements will occur, and you should not place undue reliance on these forward-looking statements. These forward-looking statements speak only as of the date hereof.We undertake no obligation to update or review any guidance or other forward-looking statements, whether as a result of new information, future developments or otherwise, except as may be required by the U.S. federal securities laws. Non-GAAP Financial MeasuresFinancial information contained in this presentation includes certain financial measures that are calculated and presented on the basis of methodologies other than in accordance with generally accepted accounting policies in the United States of America (GAAP), such as Adjusted EBITDA, which include or exclude certain items from the most directly comparable GAAP financial measure. These non-GAAP measures differ from reported GAAP measures and are intended to illustrate what management believes are relevant period-over-period comparisons and are helpful to investors as an additional tool for further understanding and assessing Innodata’s expected ongoing operating performance. A reconciliation of historical non-GAAP financial measures to the most directly comparable GAAP financial measure is included in this presentation. Any non-GAAP financial measure used in this presentation is in addition to, and not meant to be considered superior to, or a substitute for, measures prepared in accordance with GAAP.
Page 3
3 Innodata is a Global Data Engineering Company. Q4 2025 Investor Presentation Data and Artificial Intelligence (AI) are inextricably linked.Our mission is to enable the responsible advancement of artificial intelligence by providing the data, evaluation frameworks, and human expertise required to build AI systems that can be trusted at scale.
Page 4
PlatformsAgentic AI Gen AI ImplementationAI Solutions End-to-End AI Lifecycle Solutions 4 About Innodata 36+ YEARSHistory with a rich legacy as a trusted partner to leading companies. Strategically positioned as a "picks and shovels" beneficiary of the AI and GenAI revolution. Contracts with five of the “Magnificent Seven.” Contracts with leading big tech companies. We achieved 48% revenue growth in FY25
Page 5
5 5 1 Bloomberg Intelligence – Generative AI Market Size (Nov-24).2 A large language model (LLM) is a large-scale artificial intelligence system trained on extensive textual data that leverages deep learning to comprehend, generate, and work with human language effectively. Generative AI refers to a category of AI technologies, including LLMs, designed to create original content, such as text, images, music or other forms of media. Why Invest in Innodata. About Innodata The technology partner-of-choice for data quality, scale, and agility. Enormous total addressable market (TAM) opportunity for generative AI of $1.6 trillion by 2032.1 Supporting Big Tech builders and enterprise adopters of large language models (LLMs)2, agentic systems, and other AI for enterprises at scale. Proven, transferrable model that provides opportunity to expand in Big Tech and enterprise. Strong financial position: accelerated revenue growth,strong operating leverage,healthy balance sheet. Experienced leadership team and 36+ year legacy of developing large-scale, high-quality data.
Page 6
Positioned at the Data Layer of Generative AI. Launched Innodata LabsDeveloped proprietary Goldengate foundational AI platform.Used real world insights to expand addressable market and support organizations in their AI journeys. Applied AI and Machine Learning (ML) Rapidly Scaled Two Big Tech CompaniesDelivered high-quality data for training cutting-edge AI models.Created new type of training data critical for large language models (LLMs). Signed 8 Big Tech for LLM Data Engineering, Driving Significant GrowthIncludes five of “Mag Seven”.Harnessing our data capabilities and platforms to meet emerging requirements of mega-cap technology companies. Classical AI for Leading Tech CompaniesGenerative AI for Leading TechCompaniesAdvancing Frontier AI Capabilities Well-Positioned in Growing Market 2017 Onwards 2025 – Beyond2021 – 20222022 – 2024Inflection Point 6
Page 7
7 *Includes both current and prospective (roadmap) capabilities. About Innodata VERTICAL SOLUTIONS COMPETITORS AI SERVICES & PLATFORMS COMPETITORS AI Services + Platforms •Consulting•AI Services (AIOps, CoPilots, Model Trust, Safety and Evaluation)•Digital Services (Cloud, Legacy App and Product Design and Modernization) •AI Stack Services (Foundation Models, Tooling, Data Architecture, Training Data Preparation, Data Governance, Data Migration)•Business Process Management / Managed Services•Vertical PlatformsAI Scaled Data Solutions SCALED DATA SOLUTIONS COMPETITORS •Data Curation•Data Transformation•LLM Instruction Datasets / Demonstrations (Multilingual) (Supervised Fine-Tuning, RLHF, DPO)•Data Annotation (Across Modalities)•LLM Trust and Safety•Trust Layer in the Agentic Stack Innodata Supports GenAI Builders and Adopters
Page 8
8 Our Services + Platforms Supported bySubject Matter Experts Across the Globe.About Innodata •Data Science•Computational Linguistics•Linguistics•Programming 8 SME Domain Specialization Supporting customers across all time zones.Delivery Locations Subject matter experts across finance, healthcare, legal, science, and more.Global Experts Comprehensive language coverage for customer projects.Languages + Dialects •Law + Regulation•Medicine +Allied Health•Life Sciences•Physical Sciences •Engineering•Business +Finance•Retail + Ecommerce•And More…
Page 9
9 1 Our Goldengate Platform is a proprietary AI foundation model we deeply trained over past eight years. We believe it is state-of-the-art for over 50 knowledge tasks. It is at the core of certain of our engagements, tools, and enterprise deployments2. Still in Beta. Our Platforms Support Both Big Tech + Enterprise AI Use Cases.About Innodata Proprietary, State-of-the-Art Low-Code AI Platform Is Our Core AI Technology Stack Foundational Goldengate Platform1Regulatory Change Management PlatformIntegrated workflow application enabling legal and compliance teams to make informed decisions. One of world's largest banks is charter customer. LLM Annotation and Red teaming PlatformCustomized platforms for accelerating development of LLM (supervised fine tuning, RHLF) and ML (labeling, annotation) training data at scale. Media Intelligence PlatformConnect with media influencers, amplify messages, monitor coverage, and measure public relations impact. Document Intelligence PlatformExtracts intelligence from documents in accordance with a company’s proprietary taxonomy or industry taxonomy. Medical Data Intelligence PlatformExtracts detailed medical data from patient records for AI analytics, automated processing, and expert review. Our customers include several of the largest U.S. and global life insurers. Agent Evaluation & Observability Platform2Purpose built to provide pre-deployment and post-deployment of Agentic AI systems and applications.
Page 10
10 We Believe We Have Opportunities for Continued Growth +Expansion.Well-positioned in Growing Market 10 AI/LLM-Enabled Vertical PlatformsSuccessfully re-inventing the way work is done through LLM-integration/augmentation. Federal PracticeGrowing our federal AI business through differentiated data capabilities and strategic partnerships to capture meaningful opportunities. Magnificent Seven Global Tech Companies & Other Large Technology Companies Building Foundation ModelsBillions of dollars committed to LLM strategies. We support leading tech companies across pre-training, post-training, and evaluation of emerging agentic systems. Enterprise AI Services and PlatformsAssisting enterprises to integrate generative AI across products and operations. Sovereign AIStrategic engagements with international AI leaders building domestically aligned AI ecosystems.
Page 11
11 1 Magnificent Seven includes Apple, Amazon, Alphabet, Meta Platforms, Microsoft, Nvidia, and Tesla. https://sherwood.news/tech/big-tech-capex-doesnt-seem-to-be-slowing-down/2 https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/superagency-in-the-workplace-empowering-people-to-unlock-ais-full-potential-at-work.3 https://www.deloitte.com/us/en/what-we-do/capabilities/applied-artificial-intelligence/content/state-of-ai-in-the-enterprise.html 4Bloomberg Intelligence – Generative AI Market Size (Nov-24). Generative AI is Anticipated to be the Next-Generation Computation Platform.Well-positioned in Growing Market We believe the generative AI market opportunity will rival the internet and mobile in significance.•“Magnificent Seven” AI arms race now underway — billions allocated1•McKinsey & Co. believes generative AI will potentially add $4.4 trillion in added productivity growth potential from corporate use cases2•Generative AI remains a structural enterprise priority in 2026, supporting long-term demand for AI data & enablementPositioned for the next technology wave.•Generative AI is driving significant demand for high-quality data•New technology is fundamentally built on the data we provide $ in billions Enormous generative AI addressable market opportunity4 11 +46% CAGR
Page 12
12 Strategy to be the Leading Global AI Technology Company Serving Both AI Builders + AI Adopters.Well-positioned in Growing Market New Customers +Grow Revenue Engagement with Existing Customers New Capabilities +Platforms Operations + Platforms Drive growth in targeted segments. Focus Sales and Marketing efforts around priority customers. Listen to the needs of our customers and be the highest quality provider. Utilize our deep domain expertise when engaging with our customers and ensure that our teams are well-versed in our solutions. Expand addressable market by developing new offerings and capabilities. Continue to invest in our AI technology stack and in our dev/engineering capabilities. Harness the power of AI to drive continuous improvement and long-term value creation. Regularly evaluate cost structure and investments to simplify and streamline. 12
Page 13
13 LLM Training — Improving AI Outcomes Through Targeted DataRecent Innovation Areas 13 The Industry Shift As AI models grow more complex and costly to train, the focus is shifting from how much data is used to how effectively data improves model performance. Innodata Approach •Diagnose model performance gaps using structured evaluations•Engineer targeted datasets to address specific weaknesses•Validate impact before scaling Strategic Impact Positions Innodata as a partner in improving AI outcomes, not just supplying data.
Page 14
14 Enterprises require rigorous data engineering, evaluation, and assurance to deploy agentic AI at scale. Agentic AI — The Next Phase of Enterprise AIRecent Innovation Areas Agentic AI systems operate with greater autonomy, adaptability, and contextual reasoning. 14 These systems increasingly:As autonomy increases, so do: Make decisions Execute workflows Interact with complex enterprise environments Data requirements Evaluation complexity Risk, safety, and trust considerations
Page 15
15 What Innodata Does in Agentic AI TodayRecent Innovation Areas Innodata supports the development and deployment of agentic AI through four core capabilities: 15 High-quality training data and expert evaluation to build and refine agentic systems and workflows.Agentic AI Development Integration and orchestration of agents within enterprise workflows.Enterprise Agent Implementation Trace-level evaluation and performance measurement across training and production.Agentic Evaluation & ObservabilityRed teaming and adversarial simulation to strengthen agents against real-world misuse.AdversarialTesting & Resilience
Page 16
16 Physical AI — Building the Intelligence Layer for Real-World SystemsRecent Innovation Areas 16 The OpportunityAI is moving into Physical AI which includes robotics and embodied systems, creating demand for large-scale real-world data. The Data GapProgress in physical AI, particularly in world models, is constrained by the need for diverse, real-world training data. Innodata’s Advantage•Purpose-built and in-the-wild data collection capabilities that can scale from pilot to global deployment •Large-scale real world interaction datasets for robotics training •Sophisticated data efficiency tooling to ensure that models receive optimal data to achieve quality goals Early Traction•Deployed within recently announced customer engagements•Growing demand from foundation model builders and robotics innovators•Delivering measurable improvements in model accuracy and reliability
Page 17
Q4 2025 Highlights + Financial Overview. 17 Q4 2025 Investor Presentation
Page 18
1Non-GAAP measure. Please see “Reconciliation of Non-GAAP Measures” on Slide 24.2SaaS revenue represents revenue from our Agility Media Intelligence business segment and % has been calculated as of Q4’25.3Cash and cash equivalents Financial OverviewQ4 2025 Key Highlights 18 42%Adjusted Gross Margin1 22%Adjusted EBITDA as a % of revenue1 $82.2M3 Cash on Balance Sheet $72.4MQ4 2025 Revenue22%Q4 YoY Quarterly Organic Revenue Growth~92% / 8%2 Services / SaaS
Page 19
1Non-GAAP measure. Please see “Reconciliation of Non-GAAP Measures” on Slide 24.2SaaS revenue represents revenue from our Agility Media Intelligence business segment and % has been calculated for FY 20253Cash and cash equivalents Financial Overview2025 Full Year Key Highlights. 19 43%Adjusted Gross Margin1 23%Adjusted EBITDA as a % of revenue1 $82.2M3 Cash on Balance Sheet $251.7MFY2025 Revenue48%YoY Growth~91% / 9%2 Services / SaaS
Page 20
20 In Q4 2025, we delivered 22% year-over-year (YoY) revenue growth, bringing our full year revenue growth rate to 48%. In millions US$ Financial OverviewRevenue Growth. With strong business momentum, expanding customer relationships, and continued progress on our strategic roadmap, Q4’25 delivered revenue of $72.4 million, a 22% increase from the same period last year. $18.8$19.7$22.2$26.1$26.5$32.6 $52.2$59.2$58.3$58.4$62.6$72.4 Q1'23Q2'23Q3'23Q4'23Q1'24Q2'24Q3'24Q4'24Q1'25Q2'25Q3'25Q4'25
Page 21
21 1 Non-GAAP measure. Please see “Reconciliation of Non-GAAP Measures” on Slide 24. Financial OverviewQ4 2025. Historical Q1 2023 through Q4 2025In millions US$ In millions US$ Revenue increased 22% year over year to $72.4M in Q4’25, while maintaining strong Adjusted EBITDA margins, reflecting continued operating leverage. Notably, our largest customer accounted for approximately 57% of total revenue in the quarter. 18.819.722.226.126.532.652.259.258.358.462.672.4 4%8%14%16%14%9% 27%24%22%23%26%22% -10%0%10 %20 %30 % 13 .023 .033 .043 .053 .063 .073 .0 Q1'23Q2'23Q3'23Q4'23Q1'24Q2'24Q3'24Q4'24Q1'25Q2'25Q3'25Q4'25RevenueAdjusted EBITDA as a % of Revenue $72.4MQ4 2025 Revenue22%Q4 YoY Quarterly Organic Revenue Growth$15.7MQ4 2025 Adjusted EBITDA1 22%Q4 2025 Adjusted EBITDA as a % of revenue1
Page 22
22 $251.7 $340.0 20 25 20 26 FY26 Revenue Guidance of 35% or More.FY 2026 Revenue Guidance 22 In millions US$
Page 23
23 Q4 2025 Highlights and Financial Overview 1 Subject to a borrowing base limitation, with an accordion feature that enables it to expand to up to $50 million subject to the approval of Wells Fargo. We Have a Healthy Balance Sheet with High Net Cash. 23 Cash and short-term investments of cash and short-term investments as of Dec’25 Balance sheet balance sheet with no external debt/borrowings Available credit facility line of credit with Wells Fargo, no draw-downs to date1
Page 24
24 Reconciliation of Non-GAAP Measures.Q4 2025 Highlights and Financial Overview $ in MillionsFY’23FY’24FY’25Q4’24Q4’25 Net income (loss) attributable to Innodata Inc. and Subsidiaries($0.9)$28.7$32.2$10.3$8.8 Tax Provision $1.0($4.2)$9.2$1.0$2.5Interest Expense (Income) $0.4$0.3-$1.6$0.1-$0.4Depreciation and Amortization$4.7$5.8$6.9$1.6$2.0 Severance$0.6$0.0$0.0$0.0$0.0 Stock-based Compensation$4.0$4.0$11.1$1.1$2.8 NCI $0.0$0.0$0.0$0.0$0.0Adjusted EBITDA$9.9$34.6 $57.9 $14.1 $15.7 $ in MillionsFY’23FY’24FY’25Q4’24Q4’25 Gross profit attributableto Innodata Inc. and Subsidiaries$31.3$67.1$99.5$26.8$27.7 Depreciation and Amortization$4.6$5.7$6.8$1.6$2.0 Severance$0.3$0.0$0.0$0.0$0.0 Stock-based Compensation$0.3$0.3$1.7$0.1$0.4 Adjusted Gross Profit$36.5$73.1$108.0$28.4$30.1 Gross Margin %36%39%40%45%38%Adjusted Gross Margin %42%43%43%48%42%
Page 25
Appendix. 25 Q4 2025 Investor Presentation
Page 26
26 Data Engineering Services are at the Core of Generative AI and LLMs.What We Do Raw InternetTrillions of words (low-quality / huge quantity)DemonstrationsIdeal LLM prompts & responses (highest-quality / large quantity) Comparisons(Highest-quality / large quantity)Prompts(Highest-quality / large quantity) Language ModelingPredict the next tokenLanguage ModelingPredict the next tokenBinary ClassificationPredict rewards consistent with preferencesReinforcement LearningGenerate tokens that maximize the reward Base ModelSFT ModelRM ModelRL Model Evaluation + Red TeamingToxicity, hallucination, bias, alignmentDataset Algorithm Model PretrainingSupervised Fine-TuningReward ModelingReinforcement LearningEvaluation Mitigation 26
Page 27
27 We’re Helping Capture andEnrich Application Event Data for AI-Powered Automation.Case Study 1: Global Software Company | Agentic AI 27 What We Provide•Capture of digital interactions using the customer's recording agent, such asExcel actions.•Generate event logs in the process discovery platform for each command action, covering multiple examples and performing each action in as many ways as possible.•Export events from the platform to create a labeled dataset.•Enrich event data by tagging objects, fields, actions, and groups. ResultsThe customer received a comprehensive labeled dataset based on captured digital interactions, event logs, and enriched event metadata. This enabled the customer to improve the quality and coverage of their RPA software’s ability to interpret user actions across key platforms like MS Excel, Salesforce, and SAP. The ChallengeA US-based global software company that develops robotic process automation (RPA) softwareneeded to:•Record and label actions for software applications specified by the customer. The initial list of applications included MS Excel, Salesforce, and SAP.•Write natural language prompts based on identified actions and attributes of those actions for the software applications.
Page 28
28 We’re Helping SafeguardMultimodal LLMs.Case Study 2: Leading Cloud Provider | Multimodal Red Teaming Services 28 What We Provide•Expert Red Teaming with thousands of multimodal prompts created and tested monthly in four language varieties.•Continuous delivery of custom prompts and results to the customer to stress-test model behavior and identify vulnerabilities. Results•Successfully uncovered thousands of vulnerabilities across the customer’s models, including issues with severe risks that could have caused liability or brand damage.•Delivered ongoing insights and data to support model retraining and improvement, ensuring safer and more reliable model behavior prior to public launch. The ChallengeA major cloud provider preparing to launch a suite of multi-modal LLMs required:•In-depth red teaming of new models before launch.•Multi-modal testing covering input and output across text, speech, audio, images, and video.•Four language varieties: US English, German, Mexican Spanish, and US Spanish.•High-volume prompt testing: 1,500 prompts weekly over several months across various model versions, aiming for at least 75% of prompts to “break” the models by triggering harmful or policy-violating outputs.
Page 29
29 We’re Helping Improve Model Performance Through Complex RLHF.Case Study 3: Leading Tech Company | Reward Modeling for Chatbot Summarization 29 What We Provide•Expert team assembly for analyzing requests (e.g., requests for information, summarization)•Comprehensive "reward" feedback development based on specified parameters•Three types of specialized feedback depending on the task:•Prompt authoring/response modeling to demonstrate good responses.•Dissection of responses with annotations on good or bad elementsby category.•Comparative analysis between two outputs to the same question to determine which is better.Results•Established clear evaluation criteria across key dimensions.•Delivered concrete examples for each evaluation parameter to guidemodel improvement.•Enabled more accurate reward modeling for the customer’s chatbotsummarization capabilities.•Supported the customer’s RLHF pipeline with expert human feedback on complex response characteristics. The ChallengeA leading technology customer wanted to improve their chatbot's outputs and needed to:•Enhance their model's ability to understand requests, leverage information from provided passages, and summarize it as an answer.•Evaluate responses based on specific parameters: faithfulness, helpfulness, completeness, and conciseness.•Identify what constitutes "good" answers across these critical dimensions.•Optimize their chatbot's performance for summarization use cases.The customer partnered with Innodata to implement expert analysis and complex RLHF.
Page 30
30 We’re Helping Reimagine Regulatory Change Management.Case Study 4: Top-3 Global Bank | Generative AI-Enabled Industry Platforms 30 What We ProvideInnodata automated the monitoring and aggregation of laws, rules, regulations, as well as bulletins and other change notices across more than 2,000 global issuing authorities. We designed an AI-based system and fine-tuned a generative AI transformer model to identify citations within documents as well as match those citations against the bank’s internal database of over 15 million citations in order to identify critical compliance obligations. We combined the end-to-end automation capabilities into a bespoke platform we designed with the users’ input.ResultsThe customer has experienced a 75% reduction of time and effort. They now utilize the platform to quickly sift through developments, automatically identify citations of interest, and explore automated matches of compliance obligations, significantly optimizing their workflows and creating more time for higher-level legal analysis. The ChallengeOne of the most critical analytical functions in financial services institutions (and other highly-regulated entities) involves tracking changes in regulation and making required changes to internal controls and process. Known as “regulatory change management” and “horizon scanning”, it typically requires large teams of legal/regulatory specialists across countries. We began thinking about how AI could augment the human experts, resulting in improved cost structures as well as improved accuracy and speed (which, in turn, could result in lower penalties imposed by regulators).One of the world’s largest and most prominent banks signed on as our charter customer for the system. They signed a 5-year subscription-based deal with us for the new system and associated data feeds. The customer had regionally-focused teams spanned global jurisdictions, manually monitoring websites for laws, rules, and regulations to stay ahead of regulatory changes. They invested significant time and effort in navigating governmental websites, assessing individual publications, and manually searching internal citations and compliance obligations for relevance. Their previous efforts at augmenting the process had not yielded acceptable results.
Page 31
Innodata(Nasdaq: INOD) is a global data engineering company. We believe that data and Artificial Intelligence (AI) are inextricably linked.Our mission is to enable the responsible advancement of artificial intelligence by providing the data, evaluation frameworks, and human expertise required to build AI systems that can be trusted at scale.We provide a range of transferable solutions, platforms, and services for Generative AI / AI builders and adopters. In every relationship, we honor our 36+ year legacy delivering the highest quality data and outstanding outcomes for our customers. CONFIDENTIAL & PROPRIETARY INFORMATIONContains proprietary information which is the property of Innodata and must not be made available to, or copied, or used by anyone outside the company without its written authorization. Copyright © Innodata 2026.