Loading organizations...
§ Private Profile · Novi Sad, Serbia
Cloud AI hardware platform developing reconfigurable chips for custom AI solutions and complex ML workloads like vector search and 3D processing.
Based in Novi Sad, Serbia, and San Francisco, California, Anari AI develops reconfigurable cloud-based artificial intelligence hardware platforms and custom chips for data-heavy applications. The company utilizes field-programmable gate array technology to provide a hardware-as-a-service model, allowing software developers to deploy domain-specific accelerators for complex machine learning workloads like graph neural networks and 3D processing without manufacturing physical semiconductors. Operating with an estimated workforce of 20 to 50 employees, the enterprise has raised a total of $5 million in venture capital funding across multiple seed rounds to scale its proprietary ThorX chip architecture. This financial backing includes investments from prominent venture capital firms and angel investors such as Earlybird Venture Capital, Acequia Capital, Tensor Ventures, and Eric Ries. Anari AI was founded in 2020 by Jovan Stojanovic, Stefan Sredojevic, and Bogdan Vukobratovic.
Anari AI has raised $5.0M across 2 funding rounds.
Anari AI has raised $5.0M in total across 2 funding rounds.
Anari AI has raised $5.0M in total across 2 funding rounds.
Anari AI's investors include Alexander Fang, Joe Costello, Sasha Ostojic, Young Sohn, Acequia Capital, Roland Manger, Sukna Ventures, Petr Ulvr, TS Ventures, Earlybird Venture Capital, Insight Partners, Moment Ventures.
Anari AI has raised $5.0M across 2 funding rounds. Most recently, it raised $3.0M Seed in January 2023.
| Date | Round | Lead Investors | Other Investors | Status |
|---|---|---|---|---|
| Jan 31, 2023 | $3M Seed | Alexander Fang | JOE Costello, Sasha Ostojic, Young Sohn, Acequia Capital, Roland Manger, Sukna Ventures, Petr Ulvr, TS Ventures | Announced |
| Apr 1, 2021 | $2M Seed | Earlybird Venture Capital | Insight Partners, Moment Ventures, TET Ventures, Uncorrelated Ventures, Y Combinator, ALI Kutay, Jason Katzer, Eric Ries, Acequia Capital, Serbian Entrepreneurs | Announced |
Anari AI is a San Francisco-based startup founded in 2020 that develops reconfigurable AI hardware platforms, enabling personalized AI chips and infrastructure for data-heavy applications.[1][2] It builds products like the ThorX chip, which delivers 100x more efficient computing than GPUs for 3D/graph data structures, and a vector acceleration engine using FPGA technology on AWS and Azure to boost vector search performance.[1][2] Targeting AI hardware users, chip designers, and businesses needing custom AI solutions, Anari solves inefficiencies in traditional AI chips by allowing one-click personalization and rapid deployment, addressing scalability issues in cloud-based data processing.[1][2][3] The company has raised $5M total, including a $3M seed round about two years ago (around 2023), and remains in seed VC stage with early traction in the Western Balkans AI ecosystem.[1][6]
Anari AI was founded in 2020 by Jovan Stojanovic, a determined entrepreneur from Novi Sad, Serbia, previously CEO of ZITARIO, alongside a global team of researchers and machine learning experts.[2] The idea emerged from recognizing gaps in AI hardware design, where rigid chips fail to meet diverse needs for vector search, 3D/graph processing, and customizable compute—pushing the team to pioneer reconfigurable platforms from scratch.[1][2] Early momentum came via a €3.3M (~$3M) raise in 2023 from investors like TS Ventures Fund, Acequia Capital, Earlybird Venture Capital, Tensor Ventures, and Entrada Ventures, marking it as a leader in Western Balkans AI funding that year and fueling development of its ThorX chip.[2][6]
Anari AI rides the AI hardware optimization wave, where surging demand for efficient vector search and graph processing in LLMs, recommendation systems, and generative AI strains GPU limits.[1][2] Timing aligns with 2023-2024 AI funding peaks in regions like Western Balkans (€33M in 2023), despite 2024 slowdowns, as cloud giants like AWS/Azure seek FPGA alternatives for cost-effective scaling.[1][6] Market forces favoring Anari include exploding data volumes and reconfigurable compute needs, positioning it to influence ecosystems by enabling faster, cheaper AI deployment for startups and enterprises, much like how Neon aids serverless databases.[1][6]
Anari AI's seed-stage momentum and ThorX edge it for expansion into broader AI infrastructure, potentially capturing share as GPU shortages persist and reconfigurable hardware matures.[1][2] Trends like edge AI personalization and hybrid cloud-FPGA stacks will accelerate its growth, with possible Series A in 2026 amid stabilizing regional funding.[6] Its influence could evolve from niche innovator to ecosystem enabler, empowering custom AI chips and boosting vector search efficiency across industries—transforming how startups like itself build the AI hardware foundation from the ground up.[2]