Fireworks AI
NativePrivateUpdated November 3, 2026Inference infrastructure/platform for LLMs at enterprise scale
- Website
- fireworks.ai
- CEO
- Lin Qiao
- Founded
- 2022
- HQ
- Redwood City, United States
- Total Funding
- $1.8B
- Employees
- 201-500
Products & Offerings
Fireworks Inference Cloud
Serverless and dedicated GPU deployments with an optimized engine for serving open LLM, vision, audio and embedding models at production scale.
Fireworks Training API and Fireworks Lab
Managed serverless and dedicated training for SFT, DPO and reinforcement learning on open models, generally available since August 2026.
Fireworks Nexus
Drop-in routing and cost-control layer that sends routine work from Claude Code, Codex and similar tools to open models, with budgets and usage visibility.
Fireworks on Microsoft Foundry
Fireworks' inference engine built into Azure's Microsoft Foundry with enterprise SLAs and no-setup onboarding, launched in public preview in March 2026.
Ember-1
Research-preview reasoning model built on Kimi K3 for coding and agentic work, using roughly 40% fewer reasoning tokens with a 1M-token context.
Executive Summary
What they do
- Runs a fast inference cloud for open-weight LLMs and multimodal models
- Offers fine-tuning, preference tuning and reinforcement learning on open models via the Training API
- Routes coding-agent traffic between open and frontier models to cut costs
Why it matters
- Over $1B in annualized revenue and 40T+ tokens served daily
- Used in production by Cursor, Uber, Shopify, Notion, DoorDash and GitLab
- Same-day access to new open models such as Kimi K3 and DeepSeek
Best fit
- Product teams moving high-volume workloads off costly frontier model APIs
- Engineering organizations fine-tuning open models into their own specialized models
- Azure customers wanting managed open-model inference through Microsoft Foundry
Watch-outs
- Credential security incident disclosed October 2026 is still under active monitoring
- Value depends on open models keeping pace with closed frontier models
- Pricing and plan terms change frequently, including recent serverless plan revisions
Latest News
AI Magazine · 7 days ago
PositiveMugglehead Investment Magazine · 8 days ago
Positivejawlah.co · 10 days ago
PositiveThe Real Deal · 12 days ago
PositiveHoodline · 12 days ago
NeutralUnlock Full Intelligence
Pro members get the complete briefing
- Complete analyst verdict
- Risk assessment
- All 6 Bluescore dimension scores
- Traction signals
- Bluescore history & momentum tracking
- Unlimited company access
Fireworks AI is one of the clear leaders in managed inference for open models. It has $1B+ ARR, demanding production customers, Microsoft Foundry distribution and a $1.5B Series D, so most buyers moving to open or custom models should shortlist it alongside Together AI and Baseten. Before signing, get a full write-up of the October 2026 credential incident and its fixes, and benchmark latency, cost and uptime on your own workloads.
Signal Provenance
Latest observation date and source for each signal category
| Signal Category | Last Updated | Source |
|---|---|---|
Funding Signals Last funding round, Funding stage | Oct 31, 2026 | Montclare research agents |
Hiring Velocity Headcount, Headcount growth | Nov 3, 2026 | Montclare research agents |
Technical Output Research papers, GitHub activity | Nov 3, 2026 | Montclare research agents |
News Sentiment News mentions, News sentiment | Nov 3, 2026 | Montclare research agents |
Partnership Signals Enterprise partnerships, Distribution partners | Nov 3, 2026 | Montclare research agents |
Revenue Indicators ARR estimate, Revenue growth | Nov 3, 2026 | Montclare research agents |
Leadership Signals CEO tenure, Founder-led | Nov 3, 2026 | Montclare research agents |
Competitive Position Category rank, Market share | Nov 3, 2026 | Montclare research agents |
Why this Bluescore?
6 dimensions, each scored 0–10 against all — companies. Each adds its weight × score to the Bluescore — see methodology