NewAsk Claude or ChatGPT about AI tools, and they can check Orca.Add Orca to your AI assistant
ToolsCompaniesAbout
Company

FriendliAI

FriendliAI develops an AI inference cloud for deploying frontier open-weight and custom AI models, offering model APIs and dedicated endpoints, plus a bring-your-own-GPU option that runs its inference stack on customers' own hardware. Its services serve organizations that need fast, reliable, and cost-efficient generative AI model serving.

Status
Active
Checked 29 Oct 2026
Based in
Unknown
Founded
2021
Source ↗
Tools listed
1

Tools from FriendliAI

  • FriendliAI

    Inference cloud for open-weight and custom AI models

    Apps
    • Web ↗

      Sign up or log in to Friendli Suite. Use it to try models in the playground and to create API keys.

    Developer access
    • API ↗

      REST API at api.friendli.ai, mostly OpenAI-compatible. Needs a Friendli API key from Suite. Model APIs base URL: https://api.friendli.ai/serverless/v1.

    • SDK / library (Python) ↗

      Install with pip install friendli. Needs a Friendli token or API key. Docs say it covers Serverless, Dedicated and Container endpoints.

    Verified options shown; other ways to use this tool are still being checked.

    Coding · Sign up · Checked

    Dedicated GPUs from $4/hourPer-hour GPU rates billed per second; Model APIs and BYOG are priced separately.

When FriendliAI launches another product, it gets its own record here.

@friendliai on X

We read everything FriendliAI posts on X and summarize what’s worth your time: launches, new features and price changes. The noise stays out.

friendliai@friendliaiNow availableNov 3View on X ↗

GLM-5.3 now available on Vercel AI Gateway

“FriendliAI is on Vercel AI Gateway. GLM-5.3, Oct 13–20: • 266 ms median TTFT, lowest of 22 providers • 19.8x faster than the provider average Prefer FriendliAI with fallbacks, route exclusively, or bring your own key. Guide 👇 friendli.ai/blog/friendlia…”friendliai@friendliaiPriceOct 29View on X ↗

Gemma 4 31B on FriendliAI priced at $0.14/$0.40 per 1M

“For teams looking to avoid the overhead of self-hosting, managed inference can be a practical alternative. For example, Gemma 4 31B is available on FriendliAI at just $0.14 per 1M input tokens and $0.40 per 1M output tokens, with no GPU infrastructure to manage. friendli.ai/models/google/…”