Based on prompt and completion tokens used
Every finetuned LLM comes with different pricing. You can check the pricing on its page. It includes the inference cost (as per the base LLM model used in finetuned LLM)
On demand finetuned LLMs
Access to all LLMs
Scale as you go
Playground
| Base LLM | Prompt Tokens Price (per Million) | Completion Tokens Price (per Million) |
|---|---|---|
| Llama-3-8B | 0.15 | 0.15 |
| Llama-3-8B-Instruct | 0.15 | 0.15 |
| Llama-3-70B | 1.0 | 1.0 |
| Llama-3-70B-Instruct | 1.0 | 1.0 |
| Base LLM | Finetuning Tokens Price (per Million) |
|---|---|
| <= 8B | 0.3 |
| > 8B and <= 21B | 0.6 |
| > 21B and <= 34B | 1 |
| > 34B and <= 70B | 2 |
| > 70B and <= 140B | 4 |