Replicate LLM API Pricing
replicate.com43
Models
$0.085
Cheapest/1M
$2.98
Avg Price/1M
0
Free Models
What is Replicate's cheapest model?
Replicate's cheapest LLM is Ibm Granite Granite 3.3 8B Instruct at $0.085 per million tokens (blended). They offer 43 models total.
All Replicate Models by Price
16 models| # | Model | Input | Output |
|---|---|---|---|
| 14 | Gemini 3 Pro | $2.00 | $12.00 |
| 12 | Claude 4.5 Haiku | $1.00 | $5.00 |
| 15 | Claude 4.5 Sonnet | $3.00 | $15.00 |
| 16 | Xai Grok 4 | $7.20 | $36.00 |
| 11 | O4 Mini | $1.00 | $4.00 |
| 13 | GPT 4o | $2.50 | $10.00 |
| 1 | Ibm Granite Granite 3.3 8B Instruct | $0.030 | $0.250 |
| 2 | Llama 3 8B | $0.050 | $0.250 |
| 8 | Llama 3 70B | $0.650 | $2.75 |
| 5 | Mistralai Mistral 7B V0.1 | $0.050 | $0.250 |
| 3 | Llama 2 7B | $0.050 | $0.250 |
| 4 | Llama 2 7B Chat | $0.050 | $0.250 |
| 6 | Llama 2 13B | $0.100 | $0.500 |
| 7 | Llama 2 13B Chat | $0.100 | $0.500 |
| 9 | Llama 2 70B | $0.650 | $2.75 |
| 10 | Llama 2 70B Chat | $0.650 | $2.75 |
Other Providers
Frequently Asked Questions
Replicate offers 43 models tracked on Sector HQ. The cheapest model starts at $0.085 per million tokens (blended rate).