Pika

Creates Pika 2.5 video generation model optimized for speed, social media content, and rapid iteration. Integrated into Adobe Premiere Pro; known for Scene Ingredients feature and expressive, playful video style.

Fastest generation times (30-90 seconds) among major video models at accessible pricing, ideal for high-volume social creators.

pika.art Foundation Models - Video

United Statesprivate

At a glance

Total raised

$135M

last round series b

Headcount

51-200

approximate · public range

Headquarters

Palo Alto

United States

Founded

2023

3 years active

Status

Private

Foundation Models - Video

Latest valuation

$700M

Series B · 2024

For subscribers

Subscribe to see the full profile

Including Funding rounds, Bull / Bear thesis, Stock + earnings, Roster changes, Patents, News, and Open roles.

Founding

50% off

/month

94 of 100 spots left

Full

$10

/month

Available once all 100 Founding Member spots are claimed.

Get full access

Already subscribed? Sign in →

Funding history

Total raised

$135M

Rounds tracked

Latest stage

Series B

Latest valuation

$700M

Date	Stage	Amount	Valuation	Lead investors
Jun 4, 2024	Series B	$80M	$700M	Spark Capital
Nov 28, 2023	Series A	$35M	$200M	Lightspeed Venture Partners
Sep 1, 2023	Seed	$5M	—	—
Jan 1, 2023	Pre-Seed	$15M	—	Nat Friedman, Daniel Gross

Founder backgrounds

Chenlin Meng

Co-Founder & CTO

Dropped out of Stanford University's artificial intelligence PhD program in 2023 to co-found Pika with Demi Guo. Completed a BA in Mathematics at Stanford, rapidly advancing to graduate-level courses in her freshman year. Published five papers on generative AI as an undergrad under the mentorship of Professor Stefano Ermon, and authored over 30 papers in three years, with co-authorship of the DDIM paper, a key method in content generation used in OpenAI's DALLE-2, Google's Imagen, and Stability AI's Stable Diffusion. Worked at Google Brain and Stability AI before Pika. Had childhood dreams of becoming an animator, inspired by films from studios like Pixar, which directly informed her focus on building accessible, creator-friendly video generation tools.

Demi Guo

Co-Founder & CEO

Harvard graduate with a bachelor's degree in mathematics and a master's in computer science, she earned a silver medal at the 2015 International Olympiad in Informatics before pursuing a PhD at Stanford. She joined Facebook AI Research as the youngest full-time employee during her undergraduate years, and interned at Google Brain, Microsoft Bing, and Quora. During the winter of 2022-23, she and Stanford classmates participated in an AI Film Festival organized by Runway, but were frustrated with the quality of existing AI video tools. Following their dropout from Stanford's PhD program in April 2023, Guo and co-founder Chenlin Meng began developing initial prototypes for Pika's AI-powered text-to-video generation system, leveraging their expertise in artificial intelligence gained from Stanford's AI Lab. She is seen as the next generation of entrepreneurs in Silicon Valley, thanks to her innate talent, decisive work ethic, and ability to understand young users' preferences quickly.

AI bull / bear

Bull case

Pika has carved out a defensible niche in the crowded video generation market by optimizing for speed and workflow integration rather than pure quality. Generation times of 30–90 seconds represent a 3–5× advantage over competitors like Runway and Luma, which matters enormously for social creators iterating on dozens of clips daily. The Adobe Premiere Pro integration is a major moat—embedding directly into the timeline where creators already work eliminates context-switching friction and creates sticky distribution through Adobe's 30+ million Creative Cloud subscribers. The Scene Ingredients feature and distinctive playful aesthetic differentiate Pika's output, enabling brand recognition in a sea of generic AI videos. At accessible pricing, Pika is positioned as the high-volume workhorse for TikTok, Instagram, and YouTube creators who need "good enough, fast" rather than "perfect, slow." If social video continues explosive growth and speed-to-publish remains paramount, Pika owns the most valuable wedge in the market.

Bear case

Pika's speed advantage is inherently temporary—model optimization and hardware improvements will compress generation times across all competitors within 12–18 months, erasing its core differentiation. The company lacks disclosed funding, raising questions about its ability to compete in the brutal capital arms race required for foundation model development. OpenAI (Sora), Meta (Movie Gen), Google (Veo), and well-funded startups like Runway have orders of magnitude more resources for compute, data, and talent acquisition. The "expressive, playful" style risks becoming a liability if customers demand more versatile outputs for different content types. Adobe integration, while valuable, is non-exclusive—Adobe will inevitably integrate multiple video models, commoditizing Pika's distribution advantage. Social media platforms themselves (Meta, ByteDance) are building native video generation tools, threatening to cut out third-party vendors entirely. Without disclosed traction metrics or funding to sustain R&D velocity, Pika risks becoming a feature, not a company.

Generated by claude-sonnet-4-5-20250929 · May 1, 2026Refreshes quarterly · see disclosures

In the news

Pika · last 30 days

No articles ingested yet for Pika. Once the hourly news pipeline is live, every article the classifier tags as mentioning this company appears here with its one-line AI summary and sentiment.

Watch

Leadership

Demi Guo

Co-Founder & CEO

Chenlin Meng

Co-Founder & CTO

in 𝕏

Products

Pika 2.5
Pika 2.5 is the latest top-tier video generation model optimized for speed, realism, and controllability. It focuses on rapid iteration and social media content creation, delivering high-quality cinematic videos up to 1080p resolution. The model excels at text-to-video and image-to-video generation with advanced features like Pikaframes for keyframe transitions, Pikaswaps for object replacement, and creative effects. Designed for content creators, marketers, and storytellers who need fast, professional-quality video output.
Pikaformance
Pikaformance is Pika's audio-driven performance model featuring hyper-real facial expressions synced to any sound or voice. Users can make images sing, speak, rap, bark, or express emotion with near real-time generation speed. The model handles multiple languages and generates complete videos in 6 seconds or less in HD quality. Perfect for creating avatar videos, lip-sync content, and expressive character animations for social media, marketing, and entertainment.
Pika AI Selves
Pika AI Selves enables users to create persistent, portable AI agents built to their specifications. These multi-modal AI companions possess distinct personalities, persistent memory, custom voices, and visual appearances. Users can deploy their Pika Selves across platforms and social networks to collaborate on creative projects. Supports text, voice, image, and video interaction modalities, making AI agents true digital extensions of users for content creation and creative collaboration.

Open roles

View all careers ↗