Same open weights, served by 2 providers.
| Provider | Their model id | Serving | Input | Output | Cached in | Blended 3:1 | Source | |
|---|---|---|---|---|---|---|---|---|
DeepInfra Deprecated, 2025-09-11 | mistralai/Mistral-Small-3.1-24B-Instruct-2503 | standardFP8Reduced context (128k of 131k) | $0.05 | $0.10 | — | $0.063 | Verified 7h ago · Changed 9h ago | |
Google Vertex AI | mistral-small-2503 | standardQuantization not disclosedReduced context (128k of 131k) | $0.10 | $0.30 | — | $0.15 | Verified 9h 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.
| DeepInfra fp8 | $7/mo |
| Google Vertex AI | $16/mo |
Cache hits use each provider's cached-input price where it has one. Cache write costs aren't included.