Choose the plan that's right for you
Developer
Powerful speed and reliability to start your project
Business
A plan that scales with your production usage
Enterprise
Personalized configurations for serving at scale
| Base model parameter count | $/1M tokens |
|---|---|
| up to 16B | $0.20 |
| 16.1B - 80B | $0.90 |
| MoE 0B - 56B (Mixtral 8x7B) | $0.50 |
| MoE 56.1B - 176B (DBRX, Mixtral 8x22B) | $1.20 |
Per-token pricing is applied only for serverless inference. See below for dedicated deployment pricing.
LoRA models deployed to our serverless inference service are charged at the same rate as the underlying base model. There is no additional cost for serving LoRA models.
| SDXL, $/step | SDXL w/ ControlNet, $/step |
|---|---|
| $0.0002 | $0.0003 |
For image generation models like SDXL we charge based on the number of inference steps (denoising iterations).
For multi-modal models like LLaVA, each image is billed as 576 prompt tokens.
| Base model parameter count | $/1M input tokens |
|---|---|
| up to 150M | $0.008 |
| 150M - 350M | $0.016 |
Embedding model pricing is based on the number of input tokens processed by the model.
| Model | $ / 1M tokens in training |
|---|---|
| Models up to 16B parameters | $0.50 |
| Models 16.1B - 80B | $3.00 |
| Mixtral 8x7B | $2.00 |
Fireworks charges based on the total number of tokens in your fine-tuning dataset (dataset size * number of epochs).
Usage limits
Monthly usage quota is based on past spending. Adding prepaid credits will increase your total historical spend.
| Tier | Usage Limit | Qualification |
|---|---|---|
| Tier 1 | $50 / month | Default (payment method required to use more than available credits) |
| Tier 2 | $500 / month | Total historical spend of $100+ |
| Tier 3 | $5,000 / month | Total historical spend of $1,000+ |
| Tier 4 | $50,000 / month | Total historical spend of $10,000+ |
| Custom | Contact us at [email protected] | |
Dedicated deployments are billed by GPU-second. We charge $3.89 per hour for one NVIDIA A100 80 GB GPU. Pricing scales linearly when using multiple A100 GPUs. NVIDIA H100 80 GB GPUs are available to Enterprise accounts - contact us for details.