Fireworks AI

Fireworks AI is an AI model provider.Tokenando tracks 12 Fireworks AI models, with input pricing from $0.0400/M and an average blended cost of $1.48/M. Its flagship model is Qwen2.5-72B (fw).

Serverless open-model inference focused on low latency. Function calling, JSON mode and LoRA hosting.

Founded 2022HQ Redwood City, USAWebsite ↗Official pricing ↗API docs ↗

ModelsUSD per million tokens. Blend is 70% input, 30% output, our calculation.

MODELS TRACKED
12
4 categories
FLAGSHIP
Qwen2.5-72B (fw)
Via API
MIN INPUT
$0.0400/M
cheapest model in family
AVG BLENDED
$1.48/M
across 12 priced models
MAX CONTEXT
1,048,576
largest window in family

Frequently Asked Questions

How many models does Fireworks AI offer?

Tokenando tracks 12 Fireworks AI models.

How much do Fireworks AI models cost?

Fireworks AI model input pricing starts at $0.0400 per million tokens, with an average blended cost of $1.48 per million across the 12 priced models we track.

What is Fireworks AI's flagship model?

Fireworks AI's flagship model is Qwen2.5-72B (fw). It is the highest-tier Fireworks AI model we track, with input pricing of $0.9000 per million tokens.

What model categories does Fireworks AI cover?

Fireworks AI covers 4 categories: frontier, reasoning, multimodal and efficient.

Terms used on this page