Who sells this model, where, and at what price?
Pick a model and see what every cloud charges for it, side by side: per token, on a dedicated cluster, or self-hosted on a GPU instance the model actually fits on. Each price is the cloud's own public list price.
List price per million tokens, by model and cloud
| Model | AWS | Azure | OCI | Core42 | Alibaba Cloud | Google Cloud |
|---|---|---|---|---|---|---|
| OpenAI gpt-oss 120B | $0.075 in / $0.30 out | $0.15 in / $0.60 out | $0.15 in / $0.60 out | $0.15 in / $0.37 out | — | $0.09 in / $0.36 out |
| OpenAI gpt-oss 20B | $0.035 in / $0.15 out | $0.07 in / $0.30 out | $0.07 in / $0.30 out | $0.10 in / $0.30 out | — | $0.07 in / $0.25 out |
| Llama 3.3 70B | $0.72 in / $0.72 out | $0.71 in / $0.71 out | $0.72 in / $0.72 out | — | — | $0.72 in / $0.72 out |
| Llama 4 Maverick 17B | $0.24 in / $0.97 out | $0.25 in / $1.00 out | $0.72 in / $0.72 out | — | — | $0.35 in / $1.15 out |
| Llama 4 Scout 17B | $0.17 in / $0.66 out | — | $0.72 in / $0.72 out | — | — | $0.25 in / $0.70 out |
| Qwen3 32B | $0.075 in / $0.30 out | $0.30 in / $1.20 out | — | — | $0.16 in / $0.64 out | — |
| Cohere Command A | — | $0.80 in / $3.20 out | $6.24 in / $6.24 out | — | — | — |
| DeepSeek V3.1 | $0.29 in / $0.84 out | — | — | — | — | $0.60 in / $1.70 out |
| Llama 3.1 405B | $2.40 in / $2.40 out | — | $10.68 in / $10.68 out | — | — | — |
| Mistral Small | $1.00 in / $3.00 out | — | — | $0.12 in / $0.36 out | — | — |
| GPT-4o | — | $0.60 in / $5.00 out | — | $2.50 in / $10.00 out | — | — |
| GPT-5 | — | $0.05 in / $0.40 out | — | $1.25 in / $10.00 out | — | — |
| Qwen3 235B A22B | $0.11 in / $0.44 out | — | — | — | $0.287 in / $1.15 out | — |
| Amazon Nova Lite | $0.06 in / $0.24 out | — | — | — | — | — |
| Amazon Nova Pro | $0.40 in / $1.60 out | — | — | — | — | — |
| Claude Haiku 4.5 | $1.10 in / $5.50 out | — | — | — | — | — |
| Gemini 2.5 Flash | — | — | — | — | — | $0.30 in / $2.50 out |
| Gemini 2.5 Flash Lite | — | — | — | — | — | $0.10 in / $0.40 out |
| Gemini 2.5 Pro | — | — | — | — | — | $1.25 in / $10.00 out |
| Llama 3.1 70B | $0.72 in / $0.72 out | — | — | — | — | — |
| Llama 3.1 8B | $0.22 in / $0.22 out | — | — | — | — | — |
| Phi-4 | — | $0.075 in / $0.30 out | — | — | — | — |
| Mistral 7B | $0.15 in / $0.20 out | — | — | — | — | — |
| Mistral Large | $0.25 in / $0.75 out | — | — | — | — | — |
The page compares one model in full, with regions, dedicated clusters and self-hosting. Work out a monthly cost or see which hardware fits a model.