Fast, cost-efficient hosting for open-weight models.
Fireworks specialises in optimised, low-cost hosting of open-weight models — Llama, GPT-OSS, DeepSeek, Qwen, Mistral, and dozens of fine-tuned variants — across serverless and dedicated tiers. US-hosted. Competitive on price-per-token for high-throughput open-weight workloads where cost matters more than feature parity with closed frontier models.
Every model we route through Fireworks. Compare residency, ZDR, training posture, and price at a glance — full data-handling detail per route below.
| Model | Region | Zero data retention | Training | Context | Input | Output |
|---|---|---|---|---|---|---|
| US | Enterprise | No | 131K | $0.15 | $0.60 | |
| US | Enterprise | No | 131K | $0.07 | $0.30 | |
| US | Enterprise | No | 1M | $1.74 | $3.48 | |
| US | Enterprise | No | — | $0.50 | $3.00 | |
| US | Enterprise | No | 262K | $0.95 | $4.00 | |
| US | Enterprise | No | 262K | $0.60 | $3.00 | |
| US | Enterprise | No | 200K | $1.40 | $4.40 | |
| US | Enterprise | No | 197K | $0.30 | $1.20 | |
| US | Enterprise | No | 197K | $0.30 | $1.20 |
Fireworks hosts on 1 route. Each route has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.
Zero data retention is available via Opper Enterprise contract. No training on customer data. US; SCCs; DPA available.
