AI Startup Roundup: Chinese LLM Startup Raises Over $1 Billion

Every week, AI Business brings you the latest startup news.

Keep up-to-date by subscribing to the AI Business newsletter to get content straight to your inbox and listen to interviews with AI experts from Fortune 500 companies on the AI Business Podcast on Apple iTunes and Spotify.

Funding news

Moonshot AI

Beijing, China-based Moonshot AI is an AI startup developing large language models, similar to OpenAI. Last October, it unveiled its AI chatbot, Kimi Chat, built atop its own 100 billion parameter model Moonshot LLM. The chatbot has a context window of 200,000 Chinese characters.

The startup’s Chinese name is Yuezhi Anmian, according to the South China Morning Post.

Latest funding: More than $1 billion

Lead investors: Alibaba, HongShan (Chinese spin-off of Sequoia Capital in Silicon Valley)

Other investors: Whistler Capital Partners, Nashville Capital Network, Company Ventures and local angel investors

-

Abridge

Pittsburgh, PA-based Abridge creates generative AI solutions for clinical documentation in health care. Its AI-powered platform turns patient-clinician conversations into structured clinical notes in real time.

Latest funding: $150 million, series C

Lead investor: Lightspeed Venture Partners

Other investors: Redpoint Ventures, CVS Health Ventures, IVP, Spark Capital, Union Square Ventures, Bessemer Venture Partners, Wittington Ventures, Mass General Brigham Artificial Intelligence and Digital Innovation Fund (AIDIF), Kaiser Permanente Ventures

Funding plans: Research into building foundation models based on multimodal health care data.

-

Magic

San Francisco-based Magic is developing frontier-scale code models that it claims will create a true AI “coworker, not just a copilot.” On X, it tweeted that it is building an “AI software engineer.”  The startup also said it has a model with a multimillion-token context window.

“Code generation is both a product and a path to AGI, requiring new algorithms, lots of CUDA, frontier-scale training, RL, and a new UI,” the startup tweeted.

Latest funding: $117 million

Lead investors: Former GitHub CEO Nat Friedman, tech investor Daniel Gross

Other investors: CapitalG (Alphabet’s growth fund) and tech investor Elad Gil

Funding plans: To further scale up its models

-

Recogni

Recogni designs chips that enable self-driving vehicles to better perceive objects on the road. The San Jose, California-based startup said it is working on a new chip designed for “high compute, low power inference.”

Latest funding: $102 million, series C

Lead investors: Celesta Capital and GreatPoint Ventures

Other investors: Mayfield, DNS Capital, BMW i Ventures, SW Mobility Fund, Pledge Ventures and Tasaru Mobility Investments. HSBC Innovation Banking extended debt financing.

Funding plans: Funds will be used to drive next-generation system development for AI inference solutions that boost performance and power efficiency while providing the lowest total cost of ownership.

-

LangChain

LangChain is the AI startup behind the popular framework of the same name used to simplify the creation of applications using large language models.

LangChain started as an open source project before incorporating in April 2023. It went on to build solutions including Dream, a web app development platform.

Latest funding: $25 million, series A

Lead investor: Sequoia Capital

Funding plans: LangChain just launched LangSmith - a suite for large language model application development, monitoring and testing.

-

Baseimmune

U.K.-based Baseimmune uses AI to predict future pathogen mutations to generate novel vaccines.

Baseimmune is currently developing three vaccine candidates for African swine fever, coronavirus, and malaria. All are currently in preclinical development.

Latest funding: $11.4 million, series A

Lead investors: MSD Global Health Innovation Fund, IQ Capital

Other investors: Hoxton Ventures, Creator Fund, Beast Ventures, Maki.vc

Funding plans: CEO Joshua Blight said the funds will go towards accelerating technology development, scaling programs and expediting the delivery of vaccines.

-

Siftwell Analytics

Siftwell Analytics is a health care startup that uses explainable and causal AI to provide community health plans with insights about their members to drive earlier and more precise health interventions. It is based in Charlotte, North Carolina.

Latest funding: $5.8 million, seed round

Investors: AlleyCorp, Arkin Digital Health, Tau Ventures, The Charlotte Fund and angel investors

Funding plans: Funds will be used to build the team and platform, including product development and market expansion.

-

UnityAI

UnityAI, a health care tech startup based in Nashville, Tennessee, uses large language models trained by reinforcement learning to deploy AI models that optimize hospital bed management.

Latest funding: $4 million, seed round

Lead investor: Max Ventures

Other investors: Whistler Capital Partners, Nashville Capital Network, Company Ventures and local angel investors

Funding plans: Accelerate its business plans

-

Composable Prompts

Split across New York and Paris, Composable Prompts helps organizations leverage language models in their business processes and workflows.

Latest funding: $4 million, seed round

Lead investors: Elaia Partners, Illuminate Financial

Other investors: Motier Ventures, Kima Ventures, m-ai club, Super Capital

Funding plans: The company unveiled a new enterprise platform that helps businesses with governance and security when deploying large language models.