Startup profile for fal: latest funding, latest known valuation, total disclosed funding, investors, status, sources, and why the company is interesting. Part of the Venture Capital Tracker startup directory.
Venture Capital Tracker logo
Venture Capital Tracker

Startup profile · funding coverage

fal funding, valuation and investors

Real-time generative media inference APIs for image, video, audio, and 3D at scale.

Keep track of fal

Save this profile to your VCT watchlist for a quick return.

View watchlist

Funding, valuation & investors

Answer-first snapshot

Latest funding

Series D

$140M · December 2025

Latest known valuation

$4.5B (reported)

Series D · December 2025

Total disclosed equity funding

$212M

Excludes debt, grants, acquisitions, secondaries, and IPO proceeds.

Current status

private

growth · San Francisco, California

Investors in latest funding

Lead: Sequoia

Other: Kleiner Perkins , Andreessen Horowitz , Bessemer Venture Partners

Sources: latest funding Last verified: 2026-07-25

Overview

fal, founded in 2021 by Burkay Gur and Gorkem Yurtseven, provides developer infrastructure for running generative AI models on image, video, audio, and 3D workloads with emphasis on low latency and high throughput. Customers include Quora (Poe bots), Canva, and Perplexity. The San Francisco company offers managed GPU inference, model endpoints, and private deployments rather than consumer apps. Andreessen Horowitz led the seed and co-led the 2025 Series B; Sequoia led a $140M Series D in December 2025 with Kleiner Perkins, Bessemer, and NVIDIA's NVentures participating. fal reports 100M+ daily inference requests and positions itself as the "AWS for generative media."

Why fal is interesting

fal built proprietary low-latency video inference when most AI infra was batch-oriented—usage pulled three 2025 rounds (B, C, D) led by Notable/a16z then Sequoia at a reported $4.5B valuation.

Product & use cases

fal sells GPU inference infrastructure and APIs that let developers run open-source and proprietary generative models for media creation. Its stack targets real-time video generation where milliseconds of latency and cost per frame matter for production apps.

  • Powering in-app AI image and video generation for consumer products
  • Running fine-tuned or private model deployments for enterprise creative tools
  • Serving high-volume inference for chatbots and agents with multimodal outputs

Key facts

  • Series B (Feb 2025): $49M led by Notable and a16z; total funding $72M at that point (company blog)
  • Series D (Dec 2025): $140M led by Sequoia at reported $4.5B valuation (Businesswire)
  • Powers ~40% of Poe's official image/video bots per Quora CEO quote (Tech.eu)
  • 100M+ daily inference requests cited at Series B announcement (fal blog)

Funding history (newest first)

Series D

2025-12 $140M Valuation: $4.5B (reported)

Source: https://www.businesswire.com/news/home/20251209532649/en/fal-Raises-140M-in-Series-D-Led-by-Sequoia

Series B

2025-02 $49M

Source: https://blog.fal.ai/fal-raises-49m-series-b-to-power-the-future-of-ai-video/

Series A

2024 $14M
  • Kindred Ventures (lead)

Source: https://tech.eu/2025/02/12/fal-ai-secures-49m-series-b-for-ai-video-creation/

Seed

2023 $9M

Source: https://tech.eu/2025/02/12/fal-ai-secures-49m-series-b-for-ai-video-creation/

Investors in our directory

Funds linked from fal's profile — open a fund page for stage focus and related deal articles.

Andreessen Horowitz Sequoia Capital Bessemer Venture Partners Kleiner Perkins First Round

Competitive landscape

Edge: fal claims up to 10x lower latency and cost versus generic cloud inference for video workloads, with 99.99% uptime at 100M+ daily requests—purpose-built for generative media rather than general LLM hosting.

fal sits in the fast-growing generative media infra layer alongside Replicate and GPU clouds. Its bet is that video and multimodal workloads need specialized inference engines, not generic LLM hosting. Enterprise logos (Canva, Quora) and triple 2025 fundraises suggest usage traction, though exact revenue is undisclosed. Hyperscalers remain the default fallback if fal's performance premium narrows.

  • Replicate direct

    Developer-friendly model hosting API; fal emphasizes lower-latency video and larger enterprise SLAs.

  • Together AI adjacent

    Broad GPU inference for LLMs and open models; less specialized on real-time video pipelines.

  • AWS / GCP / Azure incumbent

    Hyperscaler GPU fleets; fal sells managed optimization and model marketplace on top of raw compute.

  • RunPod / Modal adjacent

    GPU cloud for ML workloads; fal differentiates on pre-optimized generative media endpoints.

Notable stories

  • Founders Burkay Gur and Gorkem Yurtseven named fal after a Turkish wordplay on speed and light—reflecting the latency obsession (Fortune, 2025).
  • fal raised three rounds in 2025 driven by usage growth rather than runway needs, per its Series D press release (Businesswire).
  • Quora CEO Adam D'Angelo publicly said fal is among the fastest-moving vendors Poe works with on inference optimization (Tech.eu).

Industries

AI & Machine Learning Developer Tools Infrastructure & Cloud

Related funding articles

Venture Capital Tracker pieces that cover fal's financing or category context.

FAQs about fal

Practical answers founders, operators, and investors typically search for.

fal provides APIs and GPU infrastructure for running generative AI models that create images, video, audio, and 3D content with low latency—aimed at developers embedding media AI in products.
Directory-mapped funds include Andreessen Horowitz (/fund/andreessen-horowitz), Sequoia (/fund/sequoia), Bessemer (/fund/bessemer-venture-partners), Kleiner Perkins (/fund/kleiner-perkins), and First Round (/fund/first-round).
Public announcements cite $72M through Series B (Feb 2025) and a $140M Series D (Dec 2025). Series C terms were not fully disclosed in sources reviewed.
Both host models for developers. fal emphasizes real-time video inference, enterprise SLAs, and proprietary optimization; Replicate is broader and often used for experimentation.
Public customers include Quora/Poe, Canva, and Perplexity per fal and Tech.eu reporting; many enterprise deployments are not named.
fal was founded in 2021 by Burkay Gur and Gorkem Yurtseven, according to company and press profiles.
Profitability has not been publicly disclosed. Multiple 2025 rounds were tied to usage growth rather than runway, per company statements.
Businesswire reported a $4.5B valuation at the December 2025 Series D. Earlier round valuations were not disclosed in sources reviewed.
fal primarily provides inference infrastructure and endpoints; it partners with model creators and labs rather than positioning as a foundation-model trainer.
a16z led fal's seed and co-led the Series B, backing generative media infrastructure as a distinct layer from application-level AI tools.
Website All startups Funding news

By Venture Capital Tracker

Last updated:

Editorial note: AI tools assisted with research, structure, or drafting. Venture Capital Tracker retains human editorial responsibility for factual accuracy, relevance, and source quality before publication.