Fireworks AI
Open-source AI models at blazing speed, optimized for your use case, scaled globally with the Fireworks Inference Cloud. Provides generative AI inference, training/fine-tuning, and deployment infrastructure.
About Fireworks AI
Open-source AI models at blazing speed, optimized for your use case, scaled globally with the Fireworks Inference Cloud. Provides generative AI inference, training/fine-tuning, and deployment infrastructure.
Products
Build prototypes, evaluate fast, and scale with confidence.
Customize model quality, speed, and costs using powerful optimization techniques including adaptive speculative decoding, custom quantization, dynamic workload handling, reinfor...
Build ultra-fast configurable voice agents with sub-500ms responses.
Deploy anywhere and scale effortlessly across 18+ global regions and 8 providers including BYOC.
The fastest and most efficient platform to serve open models.
What customers say
“Using Fireworks on Foundry, we can run repeatable, high-volume evaluations through a single Azure endpoint, which helps our team move faster from deployment to informed model decisions with more confidence.”
“why did Cursor rollout Composer 2 with @FireworksAI_HQ?”
Upwork uses Fireworks to serve custom models for Uma's tailored proposal drafts, including Llama-3.1 LoRA models. The vendor-published story reports latency nearly an order of magnitude below tested off-the-shelf…
Case studyCresta uses Fireworks to host and scale its Ocean-1 models for Knowledge Assist, serving customer-specific LoRA adapters through shared infrastructure. Fireworks reports up to 100× lower costs than GPT-4.
Case studyFireworks reports that Cursor deployed a custom Llama-3-70b model for Fast Apply using its speculative decoding API, reaching about 1,000 tokens per second and roughly 13× the speed of vanilla inference.
Case study
Security and compliance
As reported by Fireworks AI. Ask for audit reports before you buy.
- SOC 2 Type IICompliant
- HIPAACompliant
- ISO 9001Compliant
- ISO/IEC 27001Compliant
- ISO/IEC 27701Compliant
- ISO/IEC 42001Compliant
- Penetration testing performed
- Continuity and Disaster Recovery plans established
- Continuity and Disaster Recovery plans tested
- Cybersecurity insurance maintained
- Vulnerability Disclosure Program
- Employee background checks performed
- Asset disposal procedures utilized
- Network segmentation implemented
- Unique network system authentication enforced
- Remote access encrypted enforced
- Data retention procedures established
- Customer data deleted upon leaving
- Data classification policy established
- Code of Conduct acknowledged by contractors
- Control self-assessments conducted
- Internal security procedures
SubprocessorsAnara, Capy, GitHub, Inc., LangChain, Lorikeet, RingCentral, Inc., Sierra, stagewise, Understudy, emClarity, Juicer, UFO
Is this your company?
Claim the profile to keep facts current, answer buyers and submit evidence for the next ranking review.
Claim Fireworks AI