Felo Review, Pricing & Alternatives (2026)

Felo is a multilingual AI search and creation platform made by Felo Inc., a Tokyo company established in July 2024. Felo Search launched in July 2024 as a conversational search product that answers in natural language and cites web sources. The company reported a ¥1.5 billion Series A in July 2025 from Peak XV Partners and Mirae Asset Venture Investment. The Bridge reported 1.2 million users and more than 300,000 daily searches at that time. Felo's Japanese company profile lists 30 employees, including part-time staff, at the end of January 2026. The product now includes slides, agents, LiveDoc, an LLM playground, enterprise services, and an API. Its stated mission focuses on cross-language access to global information.

Felo begins with a question in the user's preferred language, searches global material, and returns a synthesized answer with source links. Users can ask follow-up questions, choose models on eligible plans, and convert research into mind maps, presentations, web pages, images, or videos. Topic collections support continuing projects, while academic search targets scholarly material and translates paper information into the user's language. LiveDoc adds a visual workspace for documents, slides, tables, and web clips, with agents using the canvas context. This combines discovery and first-draft production in one browser workspace. Its API also provides access to models and third-party services through one key and usage-based billing. Creation features use the same credit wallet, so media generation can reduce the allowance available for other tasks. Researchers and knowledge workers should pick Felo when multilingual research must quickly become structured deliverables.

Felo combines web retrieval, multilingual natural-language processing, and selectable large language models. Published product material has named GPT-5.1, Claude 4.5 Sonnet, OpenAI o3, and Gemini 3 Pro, while later release notes mention newer model availability and agent assignments. Recent updates describe Research and PPT Agents, Video Agent controls for reference uploads, aspect ratio, resolution, and duration, plus LiveDoc’s AI-assisted canvas. Felo presents its search as using multiple language models and retrieval-augmented generation. Model access varies by feature, plan, and release, so model names should not be treated as a permanent specification. Its product direction extends beyond answer generation toward agents that transform retrieved information into documents, presentations, media, and collaborative workspace content. Researchers and knowledge workers should pick Felo when multilingual discovery must lead directly to structured deliverables.

Felo uses a freemium model. The free service provides high-speed search, and the company's published comparison lists up to five Pro searches daily. Current pricing information describes one shared credit wallet for Pro Search, Research, slides, web pages, images, video, and agents. The pricing page says Pro users can receive up to 300 Pro searches daily at zero credits, while creation features pause after credits run out. Paid users can buy bonus credits that remain valid for 90 days. The current pricing page does not publish a standard Pro price, and no standard Team price is displayed. Enterprise is presented separately. Buyers should expect creation-heavy use to consume credits faster than ordinary search.

Felo’s main limitation is an unsettled public pricing picture combined with shared credits for creation tools. Perplexity is the closer search competitor, offering a more search-focused product and clearer positioning around cited answers. ChatGPT provides broader general-purpose writing and model access, while Google Gemini is a stronger fit for users already working in Google’s services. Felo is better suited to multilingual retrieval that must lead directly to a mind map, slide deck, web page, or agent workflow. It is less suitable for organizations requiring self-hosting, predictable team pricing, or unlimited media generation. Its API and enterprise services extend its business potential, but usage terms still require review. Researchers and knowledge workers should pick Felo when cross-language search and source-to-presentation work matter most.