Felo AI

What is Felo AI?

Felo AI is a multilingual multi-agent workspace that combines AI search with document generation. The free tier gives unlimited basic searches plus 5 professional searches per day; Pro is $14.99/month. Built by Felo Inc., a Tokyo startup that raised a $10.2M Series A in July 2025 led by Peak XV Partners and Mirae Asset. Key differentiators: mind-map visualization, 30+ language support with cross-language search, and — increasingly the point of the product — an agent layer that turns research into slides, documents, and video. Best for taking a research question through to a finished deliverable, and for multilingual work.

This is a meaningful shift from how Felo is usually described. It launched as an AI search engine whose party trick was visualizing answers as mind maps. In 2026 that's still there, but it's one feature among many. The home screen leads with agents — Text to Video, AI Slide, AI Designer, Social Media, Research — and the flagship addition, LiveDoc, is a document workspace, not a search feature. Felo is competing less with Perplexity now and more with the broad "AI does your knowledge work" category.

That breadth is the whole story of whether Felo is right for you. It does more than a pure search tool, and it's less focused than one.

Who is it for?

Felo suits people who need to turn research into something — a deck, a document, a summary, a translated report — rather than just get an answer and move on. The LiveDoc workflow of transcript-in, presentation-out is aimed squarely at consultants, analysts, marketers, and anyone who spends as much time formatting deliverables as producing the underlying thinking.

Multilingual users benefit most on the search side. Felo's strongest coverage is in East Asian languages — the company reports its user base is concentrated in Japan, Korea, and Taiwan — and cross-language search lets a non-English researcher query in their own language and reach English sources without switching tools.

It is not the right pick for: quick factual lookups (Perplexity, Google, or ChatGPT Search are faster and more direct), users who want one focused tool rather than a broad agent suite, comprehensive academic literature review (Undermind, Elicit, Scite), or anyone who finds multi-feature workspaces distracting rather than useful.

Key Features

• AI search with mind maps — interactive node-and-edge diagrams with multiple layouts: spider, org chart, timeline, fishbone
• LiveDoc — agentic document workspace that converts transcripts and documents into slides, web pages, summaries, or translations
• AI Slide — generates presentation decks from a prompt or source document, exports to PowerPoint, PDF, or Canva
• AI Designer — image and visual generation
• Text to Video — video generation from text
• Social Media agent — drafts and formats social content
• Agent library — prebuilt task agents including Translator, YouTube Summary, URL Summary, Fact-Checking, and Fable Research
• Multilingual search — 30+ languages with cross-language retrieval and translation inside results
• Research Mode — multi-step research for complex topics (renamed from the earlier "Deep Mode")
• Multiple AI models — routes across current frontier models including Claude Sonnet 5, Claude Opus 4.8, GPT-5.6 variants, and Grok 4.5
• Real-time social data — pulls current Reddit and X discussion alongside web sources
• Topic Collections — group threads and chat against their combined sources
• Browser extension — Chrome and Chromium browsers
• Mobile apps — iOS and Android
• Export — PDF, DOCX, Markdown, TXT, CSV; save to Notion, Google Drive, or Felo Docs
• API platform — credit-based developer access at openapi.felo.ai, compatible with Claude Code and other agent tooling

Felo AI vs Competitors 2026

Pricing checked August 2026 from each provider's official pages. Prices for third-party tools are approximate and change frequently.

Felo AI vs Perplexity: The comparison Felo is usually filed under, but the products have drifted apart — see Perplexity vs Felo. Perplexity is a focused search engine: ask, get a cited answer, done. Felo is a workspace: search is the front door, but LiveDoc, slides, and agents are where it spends its effort. For a factual lookup, Perplexity is faster and its $20 flat logic is simpler to reason about. For turning that lookup into a deck or document, Felo does it in one place. The $14.99-vs-$20 price gap is real but small enough that scope, not cost, should decide.

Felo AI vs Gamma: A more honest 2026 comparison for the document side. Gamma is a focused AI deck generator. Felo's AI Slide and LiveDoc do decks too, bundled with search and other agents. For slides as a dedicated job, Gamma is more refined. For slides as one output of a broader research workflow, Felo keeps it in one tool.

Felo AI vs Undermind: Different depth and purpose. Undermind does comprehensive scientific literature synthesis — exhaustive, citation-heavy academic research. Felo does broad, fast, multi-format knowledge work. For serious literature review, Undermind. For quick research-to-deliverable, Felo.

Pricing 2026

Pricing per felo.ai and openapi.felo.ai, checked August 2026. Annual billing saves about 16%.

Two things about this pricing are commonly misreported. First, "Standard" is the name of Felo's free tier, not a paid team plan — several review sites get this wrong. Second, Pro is not unlimited: basic searches are uncapped on both tiers, but the professional searches that trigger deeper research and agent runs are metered at 5 per day free and 100 per day on Pro. Felo's own pages are internally inconsistent here, older material citing 300 per day; the current FAQ says 100.

For active users, Pro at $14.99 is the meaningful paid tier, and the annual plan at $12.50 is the better buy once you know the workspace suits you. The free tier is genuinely usable for occasional search and light agent use.

The honest caveat on value: at $14.99 Felo is priced near focused category leaders like Perplexity at $20, while being a broader, less specialized tool. Whether that's a bargain or a stretch depends entirely on whether you use the breadth. If you live in LiveDoc and the slide agents, it's strong value. If you only ever run searches, you're paying near-Perplexity rates for a tool that's less optimized for exactly that.

Our Testing

The product has changed a lot over the past six months - there’s now a ton of additional functionality, and sometimes it’s easy to get confused about Felo’s positioning. But I really enjoyed using the documents feature, as well as the agents, which are available in a separate module. Felo Agent template library showing Translator, YouTube Summary, URL Summary, Fact-Checking, and Fable Research agents. My personal opinion is that the product needs to highlight one particular function and stand out from other competitors.

Use Cases

Meeting transcript to presentation: Drop a raw meeting transcript into LiveDoc and convert it to a formatted strategy deck in one action — the workflow shown in Felo's own onboarding. Aimed at anyone who turns calls into shareable summaries.

Research to slide deck: Run a research query, then push the findings straight into AI Slide for a first-draft presentation, without exporting to a separate deck tool.

Multilingual research: Search in your own language and reach English-language sources through cross-language retrieval — Felo's strongest structural advantage over English-first competitors, and a fit for its East Asian user base.

Topic exploration with mind maps: Enter an unfamiliar subject and use the mind map to see how concepts connect before drilling into specifics — the original Felo use case, still useful for the first hour in a new domain.

Task agents: Use prebuilt agents — translate a document, summarize a YouTube video or URL, fact-check a claim — without building a workflow from scratch.

Our Verdict

Felo AI in 2026 is a broader, more ambitious product than the "AI search with mind maps" it's often described as. The addition of LiveDoc, AI slides, video, and an agent library turns it into a research-to-deliverable workspace, and for people whose work is exactly that — research that has to become a document or deck — the breadth is genuinely useful and actively developed.

The honest considerations follow from the same breadth. Felo is less focused than a dedicated search tool, and factual-lookup speed trails Perplexity. Pro's professional searches are capped rather than unlimited. And at $14.99 it sits close to more established rivals, so the value case depends on actually using the wider feature set rather than just the search box. For a user who only wants fast answers, Perplexity or free Google AI Mode remain the better, simpler choices.

For research-to-document workflows and multilingual work, recommend. For pure factual search, the focused alternatives serve better.

Note: Felo AI does not have an affiliate program with AIVario. We earn no commission from sign-ups, and this rating carries no commercial incentive.

Best for: Turning research into slides and documents (LiveDoc, AI Slide), multilingual work across East Asian languages, topic exploration via mind maps, users who want search and document generation in one tool Not ideal for: Fast factual lookups (use Perplexity or free Google AI Mode), users who prefer one focused tool over a broad agent suite, comprehensive academic literature review (use Undermind or Elicit) Bottom line: A multi-agent research-and-document workspace that's outgrown the "Perplexity alternative" label. Strong if you use the breadth to turn research into deliverables; overkill and slightly pricey if you only need search.

Related Tools

• Perplexity — focused AI search for fast factual lookups; the simpler alternative
• Perplexity vs Felo — head-to-head on scope, price, and which fits your work
• Gamma — dedicated AI deck generation, the focused alternative to Felo's slide agents
• Undermind — deep research for comprehensive scientific literature synthesis
• Notion AI — for capturing and organizing what comes out of Felo
• NotebookLM — grounded research over your own uploaded documents