High-speed inference platform for open-source and fine-tuned models with tiered token pricing
| Tier | Price | Includes | Overage |
|---|---|---|---|
|
Serverless
usage based
|
Free
|
— |
tokens (models <4B): $0.1/1,000,000 per tokens
tokens (models 4B-16B): $0.2/1,000,000 per tokens
tokens (models 16B+): $0.9/1,000,000 per tokens
tokens (MoE 0-56B): $0.5/1,000,000 per tokens
tokens (MoE 56B+): $1.2/1,000,000 per tokens
|
|
Enterprise
custom
|
Contact Sales
|
— | — |