| Base model parameter count | $ / 1M input tokens |
|---|---|
| up to 150M | $0.008 |
| 150M - 350M | $0.016 |
| Qwen3 8B | $0.1 |
Supervised and preference fine tuning is priced per 1M training tokens. Reinforcement fine tuning jobs are priced per GPU hour (billed per second), at the same price as Fireworks on-demand deployments (see on-demand pricing below).
| Base Model | LoRA SFT | LoRA DPO | Full Param SFT | Full Param DPO |
|---|---|---|---|---|
| Models up to 16B parameters | $0.50 | $1.00 | $1.00 | $2.00 |
| Models 16.1B - 80B | $3.00 | $6.00 | $6.00 | $12.00 |
| Models 80B - 300B (e.g. Qwen3-235B, gpt-oss-120B) | $6.00 | $12.00 | $12.00 | $24.00 |
| Models >300B (e.g. DeepSeek V3, Kimi K2) | $10.00 | $20.00 | $20.00 | $40.00 |
Attach to a shared, always-on trainer pool for LoRA training on the launch models. There's no provisioning and no idle cost. You pay only for the tokens you prefill, sample, and train.
| Base Model | Context | Prefill / 1M | Cached Prefill / 1M | Sample / 1M | Train / 1M |
|---|---|---|---|---|---|
| Qwen 3.5 9B | 65K | $0.66 | $0.132 | $1.995 | $1.463 |
| Qwen 3.6 27B | 65k | $1.86 | $0.372 | $5.595 | $4.103 |
| Kimi K3 | 192k | $10.87 | $2.17 | $27.11 | $32.55 |
Dedicated Training API jobs are priced per GPU hour. Please see the On-Demand Pricing section below for details on Dedicated Training API Pricing.
| GPU Type | Price ($) per hour |
|---|---|
| H100 80 GB GPU | $7.00 |
| H200 141 GB GPU | $7.00 |
| B200 180 GB GPU | $10.00 |
| B300 288 GB GPU | $12.00 |
