Table of Contents
- Eleven v4 and Eleven v4 Turbo Investment Report
Eleven v4 and Eleven v4 Turbo Investment Report
Category: Artificial Intelligence, Audio, Developer Tools
Company Stage: Late-stage private AI platform; product-line launch
Founder or Founders: Mati Staniszewski (CEO) and Piotr Dabkowski
Headquarters: London is the company’s European headquarters; global operations
Funding: $500 million Series D at an $11 billion valuation in February 2026; later $300 million employee tender at $22 billion
Business Model: Creator subscriptions, usage-based API, and enterprise voice-agent contracts
Product Hunt Launch Date: 2026/10/02
Report Date: 2026-10-08
| Investment Metric | Assessment |
|---|---|
| Venture Potential | 88/100 |
| Unicorn Path | Clear |
| Valuation Attractiveness | Expensive |
| Evidence Confidence | 76/100 |
| Final Decision | Pass |
Executive Summary
Eleven v4 is ElevenLabs’ expressive text-to-speech model; v4 Turbo is its lower-latency version. Both support multi-speaker dialogue, emotional delivery, voice cloning, and 90-plus languages across creator products, agents, and API. (Eleven v4 announcement; model documentation)
Artificial Analysis ranked v4 Turbo first in its September 2026 Provider Voice Arena snapshot (1332±19 Elo, 1,689 votes). ElevenLabs separately reports a ~75% listener preference in its own blind tests. These support quality, not model revenue or retention. (Artificial Analysis; ElevenLabs) (Artificial Analysis; ElevenLabs announcement)
ElevenLabs reported $500 million-plus ARR by April 2026, up from $350 million at 2025 year-end, and said enterprise was 55% of revenue. These are unaudited company figures. On September 30, it announced a $300 million employee tender at a $22 billion valuation. (ARR; tender) (ARR announcement; tender announcement)
The model is a product line, not a separate security. At the tender price, ElevenLabs is valued at roughly 44x its reported spring ARR run rate using figures from different dates. Decision: Pass at the indicated price despite strong product and company quality.
Product Overview
V4 emphasizes emotion, pacing, character, and multi-speaker dialogue; Turbo targets conversational latency. Both are available in ElevenAgents, ElevenCreative, and ElevenAPI. The company lists 90-plus languages and about 100 ms median Turbo model latency; real-world latency also depends on network and application. (Model docs) (Eleven v4 announcement; model documentation)
Customers range from developers and content studios to localization teams and enterprises. V4 can support narration, dubbing, and voice agents; value depends on natural delivery, language quality, rights, and cost.
Founder and Team Assessment
ElevenLabs was founded in 2022 by CEO Mati Staniszewski and Piotr Dabkowski. The company says Staniszewski previously deployed technology for enterprises and governments at Palantir. It reported over 800 employees in September and identifies London as its European HQ. (Series D; London HQ; tender) (Series D announcement; London office announcement; tender announcement)
The team expanded from TTS into transcription, dubbing, music, and agents. Execution is strong; model research and safety remain core dependencies.
Founder Assessment: Strong founder-market fit and demonstrated research-to-product execution; governance, hiring efficiency, and model-level accountability merit diligence at this valuation.
Market Opportunity
Initial customers are developers, media teams, and enterprises deploying voice agents; expansion includes localization, games, accessibility, and multilingual service. The company reports 90-plus languages and 55% of revenue from enterprise. (Tender update) (Model documentation; tender announcement)
The platform’s reported $500 million-plus ARR demonstrates a large monetized market, but is not attributable to v4. Customer count, ACV, and model-level revenue are not public.
Traction and Growth Signals
Artificial Analysis ranked Turbo first in September; Product Hunt showed 173 points and #5 daily. The ranking supports model quality, while Product Hunt reflects launch interest, not adoption. (Artificial Analysis; Product Hunt) (ElevenLabs; Artificial Analysis; Product Hunt)
Company reports include $350 million ARR at 2025 year-end, over $500 million by April 2026, and 15 million agent conversations per week in September. Audited statements, margins, and cohort retention are not public. (ARR; tender) (ARR announcement; tender announcement)
Traction Assessment: Exceptional reported company growth; v4-specific usage, revenue, and retention remain unreported.
Competitive Position
Competitors include Cartesia Sonic 3.6, Google Gemini, OpenAI, xAI, and Inworld. The September Artificial Analysis leaderboard supports Turbo’s benchmark quality; company-run preference tests are separate evidence and neither guarantees production performance. (Artificial Analysis; ElevenLabs announcement; Cartesia Sonic)
Advantages include voice quality, creator and developer products, API reach, and enterprise agents. Custom voices and integrated workflows may raise switching costs, though APIs are easy to test across providers and larger platforms can bundle TTS.
“If the largest platform in this market launched the same feature within six months, why would customers continue using ElevenLabs?” The answer must be sustained voice quality, reliable tools and integrations, customer-owned workflows, and proven agent performance—not benchmark rank alone.
Defensibility Assessment: Medium to High
Business Model and Economics
Revenue comes from subscriptions, API credits, and enterprise contracts. Public plans range from free to paid monthly tiers, with custom enterprise pricing. Artificial Analysis lists $40 per million characters for Turbo; verify current API terms. (Pricing; Artificial Analysis) (ElevenLabs pricing; Artificial Analysis)
V4 may drive upgrades and API usage, but inference, GPU, free credits, and enterprise deployment costs affect margins. Model-level gross margin is undisclosed.
Unicorn Path
The company’s $22 billion employee-tender valuation already exceeds the $1 billion unicorn threshold. For context, at an illustrative 10x annual revenue multiple, $1 billion would correspond to $100 million in ARR. ElevenLabs’ reported $500 million-plus ARR is company-wide, not model-specific. The relevant investment question is continued growth and cash generation from the platform, not whether v4 alone could become a unicorn.
Unicorn Path: Clear (already achieved at company level)
Valuation Assessment
ElevenLabs announced a $500 million Series D at $11 billion in February and a $300 million employee tender at $22 billion in September. Against its reported $500 million-plus spring ARR, the tender implies roughly 44x ARR; this is a cross-date indicative ratio, not a primary-round multiple. (Series D announcement; ARR announcement; tender announcement)
Valuation Attractiveness: Expensive. Reported growth is exceptional, but 44x the disclosed ARR leaves little room for slower growth or margin pressure. Margin, burn, forward growth, and primary terms are needed to justify the tender price.
Key Risks
- High entry price: The indicative tender-to-ARR ratio is about 44x using figures from different dates.
- Model commoditization: Major AI platforms and specialist voice companies can rapidly improve TTS.
- Unverified unit economics: Compute costs and margins by model are not public.
- Voice rights and likeness: Cloning creates consent, copyright, and reputational exposure.
- Benchmark-to-production gap: Leaderboard scores may not predict performance across languages, environments, or long-form workloads.
- Enterprise concentration and execution: Enterprise growth brings longer deployments, support obligations, and customer concentration risks.
- Product-level opacity: v4’s contribution to revenue, retention, and cross-sell is not disclosed.
Final Assessment
Venture Potential: 88/100
| Category | Score |
|---|---|
| Market Size and Expansion Potential | 18/20 |
| Traction and Growth Evidence | 18/20 |
| Founder and Team | 14/15 |
| Product Strength | 9/10 |
| Distribution Potential | 14/15 |
| Business Model and Economics | 7/10 |
| Defensibility | 8/10 |
| Total | 88/100 |
Strengths are model quality, distribution, enterprise adoption, and platform breadth. Weaknesses are price, undisclosed unit economics, and no model-level revenue.
Evidence Confidence: 76/100
Product details, pricing, financing, tender value, ARR, and agent usage are public. Commercial metrics are management-reported, not audited; model-level revenue, margins, and retention are unavailable.
Final Decision: Pass
Pass at the $22 billion indicated price: this row is a model release, not a separate security. Reconsider a primary opportunity if updated ARR, margins, retention, and inference economics justify its terms.
Upgrade Conditions
- A new financing or tender price is supported by audited ARR, growth, and gross-margin data.
- V4 drives measurable paid API adoption, account expansion, and customer retention.
- Voice agents sustain enterprise renewals and scale without deteriorating margins.
- Independent testing confirms quality and latency across languages and production settings.
Downgrade Conditions
- ARR growth slows materially or the $500 million run rate fails to convert to recognized revenue.
- Competitors reach comparable quality at substantially lower cost.
- Gross margins fall as usage and agent workloads scale.
- Voice-cloning or likeness disputes cause material restrictions or customer losses.
Questions for Further Diligence
- What percentage of API and Creator usage has migrated to v4 and v4 Turbo?
- What revenue, gross profit, and incremental retention are attributable to these models?
- What are gross margins by model, including inference, free credits, and voice-cloning costs?
- How do v4’s price and paid usage compare with v3 and competing providers?
- What are p50 and p95 latency and failure rates under production load by region?
- How do independent quality scores vary by language, accent, and long-form content?
- What evidence supports reported ARR, growth, and enterprise revenue mix?
- What are enterprise gross retention, net retention, and customer concentration?
- How are voice consent, licensing, impersonation, and synthetic-audio disclosure enforced?
- What were the tender participation, share classes, liquidity preferences, and current primary-round terms?