Fireworks AI API
fireworks.ai
AI UtilitiesListed, not routed yet
Serverless and on-demand inference and fine-tuning for open models.
What it does
Inference platform for open models with serverless per-token pricing, on-demand GPU deployments and fine-tuning. Spending tiers (Tier 1-4, by credits added) set quotas and legacy postpaid monthly spend caps ($50 / $500 / $5,000 / $50,000).
In its API docs
Pricing
What Fireworks AI publishes for buying direct, as listed on its own pages.
Buying direct from Fireworks AI
- Qwen3 8B$0.1 / 1M tokens
Free tier: $1 in free credits. Larger plans through sales. Fireworks AI pricing, checked Sep 29, 2026
Docs
Fireworks AI's own pages, linked as published.
- API referencedocs.fireworks.ai/api-reference/introduction
- OpenAPI specdocs.fireworks.ai/merged.openapi.yaml
- Pricingfireworks.ai/pricing
- Rate limitsdocs.fireworks.ai/guides/quotas_usage/rate-limits
- Status pagestatus.fireworks.ai
- Websitefireworks.ai
Its OpenAPI document: 222 operations and rate limits
| Method | Path | Summary | Params | Source |
|---|---|---|---|---|
| GET | /v1/accounts | List Accounts | 5 | Spec |
| POST | /v1/auth/refresh | Refresh a session JWT before expiry with a fresh TTL. | 0 | Spec |
| POST | /v1/creditCodes:redeem | APIs for credit codes. Redeem Credit Code | 0 | Spec |
| POST | /v1/validateModelConfig | Validate Model Config | 0 | Spec |
| GET | /v1/accounts/{account_id} | Get Account | 2 | Spec |
| GET | /v1/accounts/{account_id}/accountUsageFilterOptions | Distinct filter values for an account/time range (serverless + dedicated; for FE, separate -yet mirrors GetAccountUsage). | 4 | Spec |
| GET | /v1/accounts/{account_id}/auditLogs | List User Audit Logs | 9 | Spec |
| GET | /v1/accounts/{account_id}/batchInferenceJobs | List Batch Inference Jobs | 6 | Spec |
| POST | /v1/accounts/{account_id}/batchInferenceJobs | Create Batch Inference Job | 2 | Spec |
| GET | /v1/accounts/{account_id}/batchInferenceJobs/{batch_inference_job_id} | Get Batch Inference Job | 3 | Spec |
| DELETE | /v1/accounts/{account_id}/batchInferenceJobs/{batch_inference_job_id} | Delete Batch Inference Job | 2 | Spec |
| GET | /v1/accounts/{account_id}/billing/summary | Get billing summary information for an account | 4 | Spec |
| POST | /v1/accounts/{account_id}/usageCosts:query | Query grouped usage cost subtotals for an account. | 1 | Spec |
| GET | /v1/accounts/{account_id}/billingSettings | Returns the singleton BillingSettings for the given account. | 2 | Spec |
| PATCH | /v1/accounts/{account_id}/billingSettings | Updates the singleton BillingSettings for the given account. | 1 | Spec |
| GET | /v1/accounts/{account_id}/billingUsage | Get Account Usage | 7 | Spec |
| POST | /v1/accounts/{account_id}/checkpoints/{checkpoint_id}:promote | Promote a checkpoint to a model. The checkpoint is identified by account + snapshot ID; the trainer job ID is passed in the request body to | 2 | Spec |
| GET | /v1/accounts/{account_id}/clusters | List Clusters | 6 | Spec |
| POST | /v1/accounts/{account_id}/clusters | Create Cluster | 1 | Spec |
| GET | /v1/accounts/{account_id}/clusters/{cluster_id} | Get Cluster | 3 | Spec |
| PATCH | /v1/accounts/{account_id}/clusters/{cluster_id} | Update Cluster | 2 | Spec |
| DELETE | /v1/accounts/{account_id}/clusters/{cluster_id} | Delete Cluster | 2 | Spec |
| GET | /v1/accounts/{account_id}/clusters/{cluster_id}:getConnectionInfo | Get Cluster Connection Info | 3 | Spec |
| GET | /v1/accounts/{account_id}/creditRedemptions | List Credit Redemptions | 6 | Spec |
| GET | /v1/accounts/{account_id}/datasets | List Datasets | 6 | Spec |
| POST | /v1/accounts/{account_id}/datasets | Create Dataset | 1 | Spec |
| GET | /v1/accounts/{account_id}/datasets/{dataset_id} | Get Dataset | 3 | Spec |
| PATCH | /v1/accounts/{account_id}/datasets/{dataset_id} | Update Dataset | 2 | Spec |
| DELETE | /v1/accounts/{account_id}/datasets/{dataset_id} | Delete Dataset | 2 | Spec |
| GET | /v1/accounts/{account_id}/datasets/{dataset_id}:getDownloadEndpoint | Get Dataset Download Endpoint | 4 | Spec |
| POST | /v1/accounts/{account_id}/datasets/{dataset_id}:getUploadEndpoint | Get Dataset Upload Endpoint | 2 | Spec |
| POST | /v1/accounts/{account_id}/datasets/{dataset_id}:splitDataset | Gateway_SplitDataset | 2 | Spec |
| POST | /v1/accounts/{account_id}/datasets/{dataset_id}:validateUpload | Validate Dataset Upload | 2 | Spec |
| GET | /v1/accounts/{account_id}/deployedModels | List LoRAs | 6 | Spec |
| POST | /v1/accounts/{account_id}/deployedModels | Load LoRA | 2 | Spec |
| GET | /v1/accounts/{account_id}/deployedModels/{deployed_model_id} | Get LoRA | 3 | Spec |
| PATCH | /v1/accounts/{account_id}/deployedModels/{deployed_model_id} | Update LoRA | 2 | Spec |
| DELETE | /v1/accounts/{account_id}/deployedModels/{deployed_model_id} | Unload LoRA | 2 | Spec |
| GET | /v1/accounts/{account_id}/deploymentShapes | List Deployment Shapes | 7 | Spec |
| POST | /v1/accounts/{account_id}/deploymentShapes | CRUD APIs for deployment shape. Create Deployment Shape | 2 | Spec |
Showing 40 of 222 operations; the full list is in the spec.
Rate limits
Spend-tier based quotas; serverless TPM adaptive limits Without a published limit GridRouter paces conservatively (1 call per second).
Sources for this page (3)
- Fireworks pricingchecked Sep 29, 2026
- Fireworks quotaschecked Sep 29, 2026
- Fireworks AI API referencechecked Sep 30, 2026