// scorecard · 42/100
verified 1mo agoAgent-readiness11/35
Agent registration (api_key)6/9
Public API5/5
OpenAPI spec0/9
MCP server0/9
llms.txt0/3
Reliability & performance0/15
Actively maintained—
Production-ready0/4
Updated recently—
Pricing11/13
Itemized public pricing6/6
Free tier / trial4/4
Free-tier generosity1/3
Docs & DX8/12
Docs quality6/8
Public API2/2
Quickstart / examples0/2
Security & compliance8/12
SOC 26/6
ISO 270010/2
GDPR0/2
Security / trust page2/2
Openness4/10
Open source0/6
Open / un-gated API4/4
Company maturity0/3
Company size—
Years operating—
// rankings · cost per standard workload
// overview
Fireworks AI is a platform for fast inference and fine-tuning of open source generative AI models, offering serverless pay-per-token serving, on-demand GPU deployments, and multiple training options including supervised, preference, and reinforcement fine-tuning. It supports 100+ models (text, vision, embeddings, audio) with optimizations for throughput/latency and OpenAI-compatible APIs. Processes 30T+ tokens daily and includes a model library and full training-to-production workflow.
Best for: Developers and teams building production inference, agentic systems, RAG, or specialized fine-tuned models at scale with open-source economics.
// pricing
Free tier · generosity 1/5
Includes: $1 free credits
Limits: self-serve only
Serverless Inferencepay per token
- Fast/Priority tiers
- 50% cached input
- 50% batch
Fine Tuningper 1M training tokens
- LoRA SFT/DPO
- Full Param SFT/DPO
- Reinforcement FT per GPU hour
On-Demand Deploymentsper GPU hour
- H100 $7
- H200 $7
- B200 $10
// security & compliance
✓ SOC 2✗ ISO 27001✗ GDPR✓ HIPAA
// features
- Serverless Inference with pay-per-token and cached/batch discounts
- On-Demand and Reserved GPU deployments (H100/H200/B200)
- Supervised/Preference/RL fine-tuning with LoRA and full-param options
- Model library with instant access to latest OSS models (Deepseek, GLM, Qwen, etc.)
- Vision, embeddings, tool calling, structured outputs, and batch API
- Autoscaling deployments and post-training model serving