Replicate API pricing
Hosted40 language models
As of Oct 14, 2026, Replicate lists 40 language models through its API. Its cheapest language model is Llama3 8B Instruct at $0.05 input / $0.25 output per million tokens.
Language models (40)
Prices per million tokens.
replicate/anthropic/claude-3.5-haiku
- Input
- $1.00
- Output
- $5.00
- Context
- —
replicate/anthropic/claude-3.5-sonnet
- Input
- $3.75
- Output
- $18.75
- Context
- —
replicate/anthropic/claude-3.7-sonnet
- Input
- $3.00
- Output
- $15.00
- Context
- —
replicate/anthropic/claude-4-sonnet
- Input
- $3.00
- Output
- $15.00
- Context
- —
replicate/deepseek-ai/deepseek-r1
- Input
- $3.75
- Output
- $10.00
- Context
- 66K
replicate/deepseek-ai/deepseek-v3
- Input
- $1.45
- Output
- $1.45
- Context
- 66K
replicate/deepseek-ai/deepseek-v3.1
- Input
- $0.672
- Output
- $2.02
- Context
- 164K
replicate/google/gemini-2.5-flash
- Input
- $0.30
- Output
- $2.50
- Context
- —
replicate/openai/gpt-4.1
- Input
- $2.00
- Output
- $8.00
- Context
- —
replicate/openai/gpt-4.1-mini
- Input
- $0.40
- Output
- $1.60
- Context
- —
replicate/openai/gpt-4.1-nano
- Input
- $0.10
- Output
- $0.40
- Context
- —
replicate/openai/gpt-4o
- Input
- $2.50
- Output
- $10.00
- Context
- —
replicate/openai/gpt-4o-mini
- Input
- $0.15
- Output
- $0.60
- Context
- —
replicate/openai/gpt-5
- Input
- $1.25
- Output
- $10.00
- Context
- —
replicate/openai/gpt-5-mini
- Input
- $0.25
- Output
- $2.00
- Context
- —
replicate/openai/gpt-5-nano
- Input
- $0.05
- Output
- $0.40
- Context
- —
replicate/openai/gpt-oss-20b
- Input
- $0.09
- Output
- $0.36
- Context
- —
replicate/openai/gpt-oss-120b
- Input
- $0.18
- Output
- $0.72
- Context
- —
replicate/xai/grok-4
- Input
- $7.20
- Output
- $36.00
- Context
- —
replicate/meta/llama-2-70b-chat
- Input
- $0.65
- Output
- $2.75
- Context
- 4K
replicate/meta/llama-3-8b
- Input
- $0.05
- Output
- $0.25
- Context
- 8K
replicate/meta/llama-3-8b-instruct
- Input
- $0.05
- Output
- $0.25
- Context
- 8K
replicate/meta/llama-3-70b
- Input
- $0.65
- Output
- $2.75
- Context
- 8K
replicate/meta/llama-3-70b-instruct
- Input
- $0.65
- Output
- $2.75
- Context
- 8K
replicate/mistralai/mixtral-8x7b-instruct-v0.1
- Input
- $0.30
- Output
- $1.00
- Context
- 4K
replicate/openai/o1
- Input
- $15.00
- Output
- $60.00
- Context
- —
replicate/openai/o4-mini
- Input
- $1.00
- Output
- $4.00
- Context
- —
replicate/qwen/qwen3-235b-a22b-instruct-2507
- Input
- $0.264
- Output
- $1.06
- Context
- —
Claude 4.5 Haiku
replicate/anthropic/claude-4.5-haiku
- Input
- $1.00
- Output
- $5.00
- Context
- —
Claude 4.5 Sonnet
replicate/anthropic/claude-4.5-sonnet
- Input
- $3.00
- Output
- $15.00
- Context
- —
Gemini 3 Pro
replicate/google/gemini-3-pro
- Input
- $2.00
- Output
- $12.00
- Context
- —
Granite 3.3 8B Instruct
replicate/ibm-granite/granite-3.3-8b-instruct
- Input
- $0.03
- Output
- $0.25
- Context
- —
Llama 2 7B
replicate/meta/llama-2-7b
- Input
- $0.05
- Output
- $0.25
- Context
- 4K
Llama 2 7B Chat
replicate/meta/llama-2-7b-chat
- Input
- $0.05
- Output
- $0.25
- Context
- 4K
Llama 2 13B
replicate/meta/llama-2-13b
- Input
- $0.10
- Output
- $0.50
- Context
- 4K
Llama 2 13B Chat
replicate/meta/llama-2-13b-chat
- Input
- $0.10
- Output
- $0.50
- Context
- 4K
Llama 2 70B
replicate/meta/llama-2-70b
- Input
- $0.65
- Output
- $2.75
- Context
- 4K
Mistral 7B Instruct V0.2
replicate/mistralai/mistral-7b-instruct-v0.2
- Input
- $0.05
- Output
- $0.25
- Context
- 4K
Mistral 7B V0.1
replicate/mistralai/mistral-7b-v0.1
- Input
- $0.05
- Output
- $0.25
- Context
- 4K
O1 Mini
replicate/openai/o1-mini
- Input
- $1.10
- Output
- $4.40
- Context
- —
Prices in USD from LiteLLM and models.dev, checked daily; confirm on Replicate's pricing page before you rely on a number.