Fireworks AI review

Fireworks AI places second on the strength of its serving options. Every serverless request bills on input, cached input and output tokens, and you can switch between Standard, Priority, Fast, Batch and Spot modes per request on the same endpoint. Batch takes 50% off, Spot takes 30% off on best-effort capacity, and cached input is billed at half price or less with no write fee. Prices start at $0.05 per 1M input tokens, and gpt-oss-120B is $0.15 input and $0.60 output. For steady traffic, Reserved Throughput adds a throughput SLA and 99.9% uptime terms, and on-demand GPU deployments bill per GPU second with no charge for start-up time. Fireworks says it serves one endpoint across 30+ regions and is a first-party provider in Microsoft Foundry. The costs to watch are GPU rates, with an H100 at $8.00 an hour, and a 50% premium for US-only serverless on select models.

Who it is for

Production teams that want to switch between standard, priority, fast, batch and spot serving per request and keep data in specific regions.

Price

Pay per token from $0.05 per 1M input tokens; gpt-oss-120B at $0.15 input and $0.60 output per 1M; H100 deployments at $8.00 per hour.

Pay per token · From $0.05 / 1M input tokens
Quoted from the vendor pricing page on November 7, 2026.

Key features

• Standard, Priority, Fast, Batch and Spot serverless modes on the same endpoint
• Batch at 50% off and Spot at 30% off Standard
• Reserved Throughput with a throughput SLA and 99.9% uptime terms
• On-demand GPU deployments billed per GPU second that scale to zero after one hour idle
• Supervised and reinforcement fine-tuning of models up to 1T+ parameters

Strengths and weaknesses

• Strength: The most serving modes per request of any platform here
• Strength: Cached input billed at half price or less with no write fee
• Strength: One endpoint across 30+ regions with data residency options
• Weakness: H100 deployments at $8.00 an hour cost more than most rivals
• Weakness: US-only serverless carries a 50% premium on select models
• Weakness: Spot capacity can drop to zero, so it needs a fallback

Key facts

• Serverless from: $0.05 per 1M input tokens
• H100 80 GB: $8.00 per hour, $0.134 per minute
• Batch discount: 50% off input and output
• Regions: 30+

Where it ranks

Fireworks AI is No. 2 in our Best AI Inference Platforms verdict and No. 7 in Best LLM APIs.