Same open weights, served by 3 providers.
| Provider | Their model id | Serving | Input | Output | Cached in | Blended 3:1 | Source | |
|---|---|---|---|---|---|---|---|---|
Azure AI Foundry | mistral-large-3 | standardglobal endpointQuantization not disclosed | $0.50 | $1.5 | — | $0.75 | Verified 7h ago · Changed 9h ago | |
Mistral Original creator | mistral-large-2512 | standardQuantization not disclosed | $0.50 | $1.5 | $0.05 | $0.75 | Verified 9h ago · Changed 9h ago | |
AWS Bedrock | mistral.mistral-large-3-675b-instruct | standardQuantization not disclosed | $0.50 | $1.5 | — | $0.75 | Verified 9h ago · Changed 9h ago | |
Azure AI Foundry | mistral-large-3 | standarddata zone (US)Quantization not disclosed | $0.55 | $1.65 | — | $0.825 | Verified 7h ago · Changed 9h ago |
Same weights don't always mean identical outputs: quantization, context limits and serving stacks can change quality, especially for tool calling. Prices are USD per 1M tokens; hyperscaler prices are for the reference region shown.
| Azure AI Foundry global endpoint | $80/mo |
| Mistral | $80/mo |
| AWS Bedrock | $80/mo |
| Azure AI Foundry data zone (US) | $88/mo |
Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.