# Pricing

> Fireworks pricing for inference, training, and Nexus. Pay per token on Serverless, reserve throughput as you scale, or run in your own cloud. See every rate.

## Pricing that scales from idea to enterprise.

Pricing

Start on pay-per-token, reserve throughput as you scale, or run Fireworks inside your own cloud. Explore Platform for systems and apps or Nexus for code and co-work.

[Get started](https://app.fireworks.ai/signup?utm_content=pricing-page) · [Talk to an expert](https://fireworks.ai/contact)

### Platform

#### Inference

##### Deploy on your terms.

Start on shared serverless capacity, guarantee it as traffic grows, or run Fireworks inside your cloud.

###### Serverless

**Pay per token · From $0.05 / 1M input tokens**

Call any serverless model in seconds. No commitment. Usage draws down prepaid credits.

**Best for:** Launching fast and spiky traffic

   - Standard, Priority, Fast, Batch and Spot modes
   - Cached input billed at half price or less, with no write fee
   - Monthly spend limits and Auto Reload

**Commitment:** No

[Get started](https://app.fireworks.ai/signup?utm_content=pricing-page)

###### Reserved Throughput

**Reserve dollars per minute · Serverless rates, reserved by the minute**

Guaranteed throughput on serverless and on-demand models, backed by an SLA.

**Best for:** Steady production traffic

   - Throughput SLA and 99.9% uptime terms
   - Adds to your adaptive rate limits
   - Usage above your reservation can bill at standard Serverless rates, never a penalty rate

**Commitment:** Yes

[Talk to us](https://fireworks.ai/contact)

###### Fireworks Sovereign

**Contract · Talk to us**

_Private preview for Enterprise_

Fireworks-managed inference on GPUs in your cloud account or data center.

**Best for:** Data residency and existing GPU capacity

   - Inference and response handling stay inside your boundary
   - Achieve over 2x tokens per GPU dollar with our engine
   - Optional overflow to Fireworks-managed capacity

**Commitment:** Yes

[Talk to us](https://fireworks.ai/contact)

##### Pick a model. Pick a mode.

Every serverless request bills on input, cached input and output tokens. Switch modes per request, on the same endpoint.

- **Standard**: The default for most traffic.
- **Priority**: Prioritize your traffic during peak load times.
- **Fast**: 100+ tokens per second for interactive apps.
- **Ultra-fast**: **New:** Available for specific models, in preview. Talk to us.
- **Batch**: 50% off input and output for jobs that don't need an immediate answer.
- **Spot**: 30% off Standard on best-effort capacity, available for select models only. Throughput can drop to zero, so retry or fall back to Standard.

##### Highlighted Serverless rates: Standard and Priority

USD per 1M tokens.

| Model | Input | Cached input | Output | Priority (input / cached / output) |
| --- | --- | --- | --- | --- |
| Ember-1 | $3.00 | $0.30 | $15.00 | $3.75 / $0.375 / $18.75 |
| Kimi K3 | $3.00 | $0.30 | $15.00 | $3.75 / $0.375 / $18.75 |
| Kimi K3.1 Pro | $2.00 | $0.20 | $10.00 | $3.00 / $0.30 / $15.00 |
| Qwen 3.8 Max | $2.00 | $0.25 | $6.00 | $3.00 / $0.375 / $9.00 |
| GLM 5.3 | $1.40 | $0.26 | $4.40 | $1.75 / $0.325 / $5.50 |
| Muse Spark Open | $1.25 | $0.15 | $4.25 | $1.875 / $0.225 / $6.375 |
| NVIDIA Nemotron 3 Ultra (Preview) | $0.60 | $0.12 | $2.40 | $0.75 / $0.15 / $3.00 |
| DeepSeek V4.1 Flash | $0.30 | $0.006 | $1.20 | $0.375 / $0.0075 / $1.50 |
| DeepSeek V4.1 Flash 1017 | $0.30 | $0.006 | $1.20 | $0.375 / $0.0075 / $1.50 |
| MiniMax M3 | $0.30 | $0.06 | $1.20 | $0.45 / $0.09 / $1.80 |
| GLM 5.3 Flash | $0.15 | $0.03 | $0.50 | $0.1875 / $0.0375 / $0.625 |
| OpenAI GPT OSS 120B | $0.15 | $0.015 | $0.60 | $0.18 / $0.018 / $0.72 |
| NVIDIA Nemotron 3.5 Lightning 30B A3B | $0.05 | $0.01 | $0.20 | $0.0625 / $0.0125 / $0.25 |

##### Spot rates

USD per 1M tokens.

US-only Serverless is available on select models at a 50% premium (waived on Muse Spark Open); [see pricing here](https://docs.fireworks.ai/serverless/us-only-serverless). Models not listed here are priced by size.

[See all models and rates](https://docs.fireworks.ai/serverless/pricing)

| Model | Tier | Cached input | Uncached input | Output |
| --- | --- | --- | --- | --- |
| GLM-5.3 | Standard | $0.26 | $1.40 | $4.40 |
| GLM-5.3 | Spot | $0.182 | $0.98 | $3.08 |
| Kimi K3 | Standard | $0.30 | $3.00 | $15.00 |
| Kimi K3 | Spot | $0.21 | $2.10 | $10.50 |
| DeepSeek V4.1 Flash | Standard | $0.006 | $0.30 | $1.20 |
| DeepSeek V4.1 Flash | Spot | $0.0042 | $0.21 | $0.84 |
| DeepSeek V4.1 Flash 1017 | Standard | $0.006 | $0.30 | $1.20 |
| DeepSeek V4.1 Flash 1017 | Spot | $0.0042 | $0.21 | $0.84 |

##### Embeddings

| Base model | Price |
| --- | --- |
| Up to 150M parameters | $0.008 |
| 150M to 350M parameters | $0.016 |
| Qwen3 8B | $0.10 |

##### Lower the cost of every finished task.

Token price is one input. Cache hits, the mode you choose and the model each turn goes to decide the rest.

- **Keep context cached**: Session affinity keeps each agent on the server that already holds its context. Long runs reach cache hit rates up to 99.6%. · [Prompt caching](https://docs.fireworks.ai/guides/prompt-caching)
- **Batch what can wait**: Async jobs run at half the Standard rate. · [Batch inference](https://docs.fireworks.ai/guides/batch-inference)
- **Send background work to Spot**: Take 30% off for traffic that can retry or fall back · [Spot Serverless](https://docs.fireworks.ai/guides/spot-serverless)
- **Route each turn**: Send routine work to open models and save premium models for the turns that earn it.

#### Training

##### Pick your training path, from co-development with our proven researchers to running your own loop.

Train at the rates below, then serve a fine-tuned model on Serverless or a Dedicated Deployment. See our inference rates or talk to us about a custom deployment.

##### Muse Glimmer 30B SFT is free through November 30

In the UI and the Serverless Training API. No credit card required.

[Start a job](https://app.fireworks.ai/dashboard/fine-tuning/create?type=supervised&baseModel=accounts%2Ffireworks%2Fmodels%2Fmuse-glimmer-30b)

###### Serverless Training API

**By model · see table below**

Customize your own training loop

**Best for:** ML engineers and ML researchers

   - Write your loop in Python and attach to an always-on trainer pool
   - Nothing to provision, no idle cost. You pay for the tokens you prefill, sample and train
   - LoRA on popular models, including RL

[Start training](https://docs.fireworks.ai/fine-tuning/training-api/serverless)

###### Dedicated Training API

**From $8.00 / GPU-hour**

You run the loop, at full scale

**Best for:** Advanced research teams

   - Provisioned trainers and deployments
   - Full-parameter training, the largest MoE models, longer context and higher LoRA ranks
   - Billed at On Demand GPU rates

[Talk to us](https://fireworks.ai/contact)

##### Forward-deployed researchers and engineers train your model with your team, from eval design through production.

We build with you

Priced per engagement.

[Talk to Fireworks Lab](https://fireworks.ai/contact)

##### Serverless Training API

Billed per 1M tokens you prefill, sample and train. Repeated prefixes bill 80% off.

| Base model | Context | Prefill | Cached prefill | Sample | Train |
| --- | --- | --- | --- | --- | --- |
| Kimi K3 | 192K | $10.87 | $2.17 | $27.11 | $32.55 |
| Muse Spark Open | 256K | $6.19 | $1.239 | $15.48 | $18.58 |
| GLM 5.3 | 256K | $4.86 | $0.972 | $12.15 | $14.58 |
| GLM 5.3 Flash | 195K | $2.96 | $0.593 | $7.41 | $8.89 |
| Muse Glimmer 30B | 128K | $1.96 | $0.39 | $4.88 | $5.86 |
| Qwen 3.8 27B | 128K | $1.86 | $0.372 | $5.595 | $4.103 |
| DeepSeek V4.1 Flash 1017 | 256K | $0.60 | $0.12 | $3.60 | $4.98 |
| NVIDIA Nemotron 3.5 Super VL 120B | 256K | $0.57 | $0.114 | $1.44 | $2.32 |
| Qwen 3.6 35B A3B | 256K | $0.45 | $0.09 | $1.335 | $2.14 |

##### Dedicated Training API

Billed per GPU-hour at the GPU rates below.

[See GPU rates](https://fireworks.ai/pricing#gpu-rates)

#### One GPU rate card for serving and training.

On Demand bills per GPU-second and the Dedicated Training API per GPU-hour, from the same rates. No charge for On Demand start-up time.

[Talk to us about reserved capacity](https://fireworks.ai/contact)

#### GPU rates

Idle deployments scale to zero after one hour by default. Region-restricted deployments are priced at 1.5x. Reserve GPUs, typically for one year, for guaranteed capacity, higher quotas, lower hourly rates and early access to new regions and hardware.

| GPU | Per minute | Per hour |
| --- | --- | --- |
| H100 80 GB | $0.134 | $8.00 |
| H200 141 GB | $0.134 | $8.00 |
| B200 180 GB | $0.217 | $13.00 |
| B300 288 GB | $0.250 | $15.00 |
| GB300 288 GB | $0.334 | $20.00 |

#### Buy the way you already buy.

##### Azure

Fireworks is a first-party provider in Microsoft Foundry, so usage counts toward your Microsoft Azure Consumption Commitment.

##### Volume

Discounts are available for bulk and prepaid purchases. Contracted customers can move to post-paid invoicing.

##### Control

Set monthly spend limits, and tag requests by team, project and environment for chargeback.

##### Security

Zero data retention by default. SOC 2 Type II, ISO 27001, 27701 and 42001, and HIPAA with a BAA.

[Visit the trust center](https://trust.fireworks.ai/)

#### Questions we hear before the first invoice

FAQs

##### How does billing work?

You buy credits, and usage draws them down: per token for Serverless, per GPU-second for On Demand, and per token or GPU-hour for training.

##### Can I cap what I spend?

Yes. Set a monthly spend limit on your account. Auto Reload is separate: it buys more credits when your balance runs low, so production keeps running, and it never raises your limit.

##### How is prompt caching billed?

Caching is automatic, with no setup and no write fee. Cached input bills at 50% off by default, with deeper discounts on select models.

##### What happens when we go over our reservation?

We customize inference to your needs. Usage above your reservation can bill at standard Serverless rates, never a penalty rate. Overage isn't covered by the SLA and follows your Serverless rate limits.

##### When should we move from Serverless to On Demand?

When utilization is high, when rate limits get in your way, or when you want to serve a model that isn't on Serverless, including one you trained. On Demand bills per GPU-second, so a busy deployment usually costs less per token.

##### Does serving a trained model cost extra?

No. You pay for the deployment by the GPU-second, plus the cost of the training job.

##### How is Fireworks Sovereign priced?

By contract. Sovereign is available in preview to Enterprise customers. Talk to us to scope your environment.

### Nexus

#### Choose how you pay for Nexus

Save with a monthly plan, or talk to us about Enterprise.

##### Pro

**$8** / month

For individual developers.

- The right model for each task
- Open models included
- Usage against your plan cap
- Works with your AI tools

[Try Nexus](https://nexus.fireworks.ai/signup?fw_exp=nexus&utm_content=pricing-page)

##### Team

**$10** per seat / month

For teams, with usage per member.

- Everything in Pro
- Up to 200 seats
- Members, seats and billing
- Higher usage per seat

[Try Nexus](https://nexus.fireworks.ai/signup?fw_exp=nexus&utm_content=pricing-page)

##### Enterprise

**Let’s talk**

Custom terms and controls.

- SSO, SCIM and audit logs
- Groups and Model Policy
- Account, user and group limits
- Custom terms and support

[Contact sales](https://fireworks.ai/contact)

Prices in USD. Taxes may apply.

## Questions? Talk to an expert today.

Get in touch to discuss your goals, workloads, and find the best deployment option for your business.

[Talk to an expert](https://fireworks.ai/contact)