Inference provider
Replicate API pricing: every model, compared
We track 35 Replicate prices that other providers also sell. Replicate has the lowest listed price for 20 of them; for the rest, the table shows how much more you pay than at the cheapest provider.
Standard list prices in USD. Last price change Sep 28, 2026. Replicate's official pricing page. How we rank.
LLMs, most widely offered first
Replicate LLM prices vs the cheapest provider
| Model | Rank | Cheapest elsewhere | Input / 1M | Output / 1M | Context |
|---|---|---|---|---|---|
| gpt-oss-120b | #16 of 18 | W&B Inference (CoreWeave)you pay 385% more here | $0.18 | $0.72 | - |
| gpt-oss-20b | #9 of 10 | W&B Inference (CoreWeave)you pay 186% more here | $0.09 | $0.36 | - |
| DeepSeek-R1 | #7 of 8 | SiliconFlowyou pay 477% more here | $3.75 | $10.00 | 66K |
| DeepSeek-V3 | #8 of 8 | SiliconFlowyou pay 231% more here | $1.45 | $1.45 | 66K |
| DeepSeek-V3.1 | #6 of 7 | DeepInfrayou pay 137% more here | $0.672 | $2.02 | 164K |
| Qwen3 235B A22B Instruct 2507 | #5 of 6 | DeepInfrayou pay 126% more here | $0.264 | $1.06 | - |
| Gemini 2.5 Flash | #2 of 4 | Oracle Cloud Generative AIyou pay 224% more here | $0.30 | $2.50 | - |
| GPT-5 | #1 of 4 | Replicate is cheapest | $1.25 | $10.00 | - |
| GPT-5 Mini | #1 of 4 | Replicate is cheapest | $0.25 | $2.00 | - |
| GPT-5 Nano | #1 of 4 | Replicate is cheapest | $0.05 | $0.40 | - |
| llama3-70b-instruct | #3 of 4 | Novita AIyou pay 107% more here | $0.65 | $2.75 | 8K |
| llama3-8b-instruct | #3 of 4 | DeepInfrayou pay 167% more here | $0.05 | $0.25 | 8K |
| claude-3-7-sonnet | #1 of 3 | Replicate is cheapest | $3.00 | $15.00 | - |
| GPT-4.1 | #1 of 3 | Replicate is cheapest | $2.00 | $8.00 | - |
| GPT-4.1 mini | #1 of 3 | Replicate is cheapest | $0.40 | $1.60 | - |
| GPT-4o | #1 of 3 | Replicate is cheapest | $2.50 | $10.00 | - |
| GPT-4o mini | #1 of 3 | Replicate is cheapest | $0.15 | $0.60 | - |
| llama-2-70b-chat | #2 of 3 | Fireworks AIyou pay 31% more here | $0.65 | $2.75 | 4K |
| mixtral-8x7b-instruct-v0.1 | #2 of 3 | DeepInfrayou pay 19% more here | $0.30 | $1.00 | 4K |
| claude-3-5-haiku | #2 of 2 | Amazon Bedrockyou pay 25% more here | $1.00 | $5.00 | - |
| claude-3-5-sonnet | #2 of 2 | Amazon Bedrockyou pay 25% more here | $3.75 | $18.75 | - |
| claude-4-sonnet | #1 of 2 | Replicate is cheapest | $3.00 | $15.00 | - |
| gpt-4.1-nano | #1 of 2 | Replicate is cheapest | $0.10 | $0.40 | - |
| granite-3-3-8b-instruct | #1 of 2 | Replicate is cheapest | $0.03 | $0.25 | - |
| grok-4 | #2 of 2 | Microsoft Azure AI Foundryyou pay 140% more here | $7.20 | $36.00 | - |
| llama-v2-13b | #1 of 2 | Replicate is cheapest | $0.10 | $0.50 | 4K |
| llama-v2-13b-chat | #1 of 2 | Replicate is cheapest | $0.10 | $0.50 | 4K |
| llama-v2-70b | #2 of 2 | Fireworks AIyou pay 1075% more here | $0.65 | $2.75 | 4K |
| llama-v2-7b | #1 of 2 | Replicate is cheapest | $0.05 | $0.25 | 4K |
| llama-v2-7b-chat | #1 of 2 | Replicate is cheapest | $0.05 | $0.25 | 4K |
| llama-v3-8b | #1 of 2 | Replicate is cheapest | $0.05 | $0.25 | 8K |
| mistral-7b-instruct-v0.2 | #1 of 2 | Replicate is cheapest | $0.05 | $0.25 | 4K |
| o1 | #1 of 2 | Replicate is cheapest | $15.00 | $60.00 | - |
| o1-mini | #1 of 2 | Replicate is cheapest | $1.10 | $4.40 | - |
| o4-mini | #1 of 2 | Replicate is cheapest | $1.00 | $4.00 | - |
Rank and difference use the blended price (3 input : 1 output tokens). Equal prices share a rank.
Ranked for your volume
gpt-oss-120b monthly cost by provider
Change the monthly tokens and the ranking updates instantly, using each provider's standard list price.
More requirements: caching, context, features, provider
| Provider | Estimated / month | Input / 1M | Output / 1M | Cached input / 1M | Context |
|---|---|---|---|---|---|
1 W&B Inference (CoreWeave)lowest for your volumePricing page | $0.64 | $0.03 | $0.17 | Not listed | 131,000 |
2 DeepInfraPricing page | $0.71 | $0.037 | $0.17 | Not listed | 131,072 |
3 Novita AIPricing page | $1.00 | $0.05 | $0.25 | Not listed | 131,072 |
4 SiliconFlowPricing page | $1.40 | $0.05 | $0.45 | Not listed | 131,000 |
5 OVHcloud AI EndpointsBilled in EUR: €0.08 input, €0.40 output per 1M tokens. USD at the ECB rate of Oct 30, 2026 (1.1172).Pricing page | $1.79 | $0.0894 | $0.4469 | Not listed | 131,000 |
6 BasetenPricing page | $2.00 | $0.10 | $0.50 | $0.10 | 128,072 |
7 Amazon BedrockPricing page | $2.70 | $0.15 | $0.60 | Not listed | 131,072 |
8 Fireworks AIPricing page | $2.70 | $0.15 | $0.60 | $0.015 | 131,072 |
9 Microsoft Azure AI FoundryPricing page | $2.70 | $0.15 | $0.60 | Not listed | 131,072 |
10 Nebius Token FactoryPricing page | $2.70 | $0.15 | $0.60 | Not listed | 131,072 |
Input volume includes the cached share, not an additional token volume. This is a list-price estimate, not a quote: cache-write charges, cache eligibility/TTL, batch discounts, long-context surcharges, regional pricing and taxes are excluded. Check the provider's terms for your workload before switching.
Ranked #1? Show it on your site
Replicate has the lowest blended list price (3 input : 1 output tokens) for these models. The badge states that basis, updates daily and always shows the current rank, whatever it is; ranking is by price only.
<a href="https://aiselect.online/models/gpt-5"><img src="https://aiselect.online/badge/gpt-5/replicate.svg" alt="GPT-5 API price ranking on AISelect" height="20"></a>
<a href="https://aiselect.online/models/gpt-5-mini"><img src="https://aiselect.online/badge/gpt-5-mini/replicate.svg" alt="GPT-5 Mini API price ranking on AISelect" height="20"></a>
<a href="https://aiselect.online/models/gpt-5-nano"><img src="https://aiselect.online/badge/gpt-5-nano/replicate.svg" alt="GPT-5 Nano API price ranking on AISelect" height="20"></a>Compare more
Other providers
Alibaba Cloud Model Studio, Amazon Bedrock, Anthropic, Baseten, Cloudflare Workers AI, Cohere, DeepInfra, DeepSeek, fal, Fireworks AI, FriendliAI, Google Cloud Vertex AI, Google Gemini API, IBM watsonx.ai, Lambda, Microsoft Azure AI Foundry, Mistral AI, Moonshot AI, Nebius Token Factory, Novita AI, Nscale, OpenAI, Oracle Cloud Generative AI, OVHcloud AI Endpoints, Perplexity, SambaNova, Scaleway, SiliconFlow, Together AI, W&B Inference (CoreWeave), xAI, Z.ai