Text model
accounts/fireworks/routers/deepseek-v4p1-flash-1017-spot pricing
Fireworks AI charges $0.21 per million input tokens and $0.84 per million output tokens for accounts/fireworks/routers/deepseek-v4p1-flash-1017-spot, and $0.0042 for cached input. Below: every rate that applies, what common workloads cost, and cheaper options.
- Input
- $0.21/ 1M
- Output
- $0.84/ 1M
- Cached input
- $0.0042/ 1M
- Context
- 1Mtokens
All rates
| Tier | Input | Output | Cache read | Cache write |
|---|---|---|---|---|
| Standard | $0.21 | $0.84 | $0.0042 | — |
USD per million tokens. Long-context tiers apply when the whole prompt, cached tokens included, passes the threshold.
Work it out
What common workloads cost
| Workload | Per request | Cost per request | Requests a month | Per month |
|---|---|---|---|---|
| Chat reply | 1,500 input and 400 output tokens | $0.000651 | 100,000 | $65.10 |
| RAG answer, cached context | 2,000 fresh + 4,000 cached input, 500 output | $0.000857 | 100,000 | $85.68 |
| Agent step | 4,000 fresh + 8,000 cached input, 800 output | $0.00155 | 250,000 | $386.40 |
| Batch summary | 3,000 input and 300 output tokens via the Batch API | — | 1,000,000 | — |
Planning a mix of models? Simulate a workload with accounts/fireworks/routers/deepseek-v4p1-flash-1017-spot, alongside anything else you run.
Cheaper options
Current models that cost at least 15% less for a typical chat reply. Cheaper isn't always equivalent: test quality on your own prompts, then use veqen diff to see the impact at your volume.
| Model | Chat reply | Saving |
|---|---|---|
| fireworks/accounts/fireworks/models/llama4-scout-instruct-basic | $0.000465 | 29% |
| fireworks/accounts/fireworks/models/minimax-m3p1-flash | $0.000465 | 29% |
| deepinfra/deepseek-ai/DeepSeek-V3.2 | $0.000542 | 17% |
| deepinfra/Qwen/Qwen2.5-VL-32B-Instruct | $0.00054 | 17% |
Price history
| 2026-10-30 | Added to the price book |
Compare
Last verified with price book 2026-10-30. Sources: Fireworks AI pricing and LiteLLM's MIT-licensed price file. Track every request to this model, per customer or project: install the SDK.