Azure AI Foundry models & pricing
Azure AI Foundry hosts 76 models (71 with public pricing) covering 5 modalities. Azure marketplace catalog for Mistral, Cohere, Meta, Phi, and other partner models. Cheapest input starts at $0.0400/M tokens; the most premium goes up to $3,200/M. Use Future AGI's Agent Command Center to route any Azure AI Foundry model with cost-optimized fallback and unified observability.
chat 60
| Model ↕ | Input / 1M ↕ | Output / 1M ↕ | Context ↕ | Caps |
|---|---|---|---|---|
| Ministral 3B | $0.0400/M | $0.0400/M | 128,000 | tools |
| Phi 4 mini Instruct | $0.0750/M | $0.300/M | 131,072 | tools |
| Phi 4 mini Reasoning | $0.0800/M | $0.320/M | 131,072 | tools |
| Phi 4 Multimodal Instruct | $0.0800/M | $0.320/M | 131,072 | tools · vision · audio |
| Mistral Small 2503 | $0.1000/M | $0.300/M | 128,000 | tools · vision |
| Phi 4 | $0.125/M | $0.500/M | 16,384 | tools |
| Phi 4 Reasoning | $0.125/M | $0.500/M | 32,768 | tools · reasoning |
| Phi 3.5 mini Instruct | $0.130/M | $0.520/M | 128,000 | |
| Phi 3.5 Vision Instruct | $0.130/M | $0.520/M | 128,000 | vision |
| Phi 3 mini 128k Instruct | $0.130/M | $0.520/M | 128,000 | |
| Phi 3 mini 4k Instruct | $0.130/M | $0.520/M | 4,096 | |
| Model Router | $0.140/M | — | — | |
| GPT Oss 120B | $0.150/M | $0.600/M | 131,072 | tools |
| Mistral Nemo | $0.150/M | $0.150/M | 131,072 | tools |
| Phi 3 Small 128k Instruct | $0.150/M | $0.600/M | 128,000 | |
| Phi 3 Small 8k Instruct | $0.150/M | $0.600/M | 8,192 | |
| Phi 3.5 MOE Instruct | $0.160/M | $0.640/M | 128,000 | |
| Phi 3 Medium 128k Instruct | $0.170/M | $0.680/M | 128,000 | |
| Phi 3 Medium 4k Instruct | $0.170/M | $0.680/M | 4,096 | |
| Grok 4.1 Fast Non Reasoning | $0.200/M | $0.500/M | 131,072 | tools |
| Grok 4.1 Fast Reasoning | $0.200/M | $0.500/M | 131,072 | tools · reasoning |
| Grok 4 Fast Non Reasoning | $0.200/M | $0.500/M | 131,072 | tools |
| Grok 4 Fast Reasoning | $0.200/M | $0.500/M | 131,072 | tools |
| Grok Code Fast 1 | $0.200/M | $1.50/M | 131,072 | tools |
| Llama 4 Scout 17B 16e Instruct | $0.200/M | $0.780/M | 10,000,000 | tools · vision |
| Grok 3 mini | $0.250/M | $1.27/M | 131,072 | tools · reasoning |
| Grok 3 mini | $0.250/M | $1.27/M | 131,072 | tools · reasoning |
| Meta Llama 3.1 8B Instruct | $0.300/M | $0.610/M | 128,000 | |
| Llama 3.2 11B Vision Instruct | $0.370/M | $0.370/M | 128,000 | tools · vision |
| Mistral Medium 2505 | $0.400/M | $2.00/M | 131,072 | tools |
| Jamba Instruct | $0.500/M | $0.700/M | 70,000 | |
| Mistral Large 3 | $0.500/M | $1.50/M | 256,000 | tools · vision |
| DeepSeek v3.2 | $0.580/M | $1.68/M | 163,840 | tools · reasoning · cache |
| DeepSeek V3.2 Speciale | $0.580/M | $1.68/M | 163,840 | tools · reasoning · cache |
| Kimi K2.5 | $0.600/M | $3.00/M | 262,144 | tools · vision |
| Llama 3.3 70B Instruct | $0.710/M | $0.710/M | 128,000 | tools |
| Claude Haiku 4.5 | $1.00/M | $5.00/M | 200,000 | tools · vision · reasoning · cache |
| Mistral Small | $1.00/M | $3.00/M | 32,000 | tools |
| Meta Llama 3.70B Instruct | $1.10/M | $0.370/M | 8,192 | |
| DeepSeek V3 | $1.14/M | $4.56/M | 128,000 | |
| DeepSeek v3.0324 | $1.14/M | $4.56/M | 128,000 | tools |
| DeepSeek R1 | $1.35/M | $5.40/M | 128,000 | reasoning |
| Mai Ds R1 | $1.35/M | $5.40/M | 128,000 | reasoning |
| Llama 4 Maverick 17B 128e Instruct Fp8 | $1.41/M | $0.350/M | 1,000,000 | tools · vision |
| Mistral Large 2407 | $2.00/M | $6.00/M | 128,000 | tools |
| Mistral Large latest | $2.00/M | $6.00/M | 128,000 | tools |
| Llama 3.2 90B Vision Instruct | $2.04/M | $2.04/M | 128,000 | tools · vision |
| Meta Llama 3.1 70B Instruct | $2.68/M | $3.54/M | 128,000 | |
| Claude Sonnet 4.5 | $3.00/M | $15.00/M | 200,000 | tools · vision · reasoning · cache |
| Claude Sonnet 4.6 | $3.00/M | $15.00/M | 1,000,000 | tools · vision · reasoning · cache |
| Grok 3 | $3.00/M | $15.00/M | 131,072 | tools |
| Grok 3 | $3.00/M | $15.00/M | 131,072 | tools |
| Grok 4 | $3.00/M | $15.00/M | 131,072 | tools |
| Mistral Large | $4.00/M | $12.00/M | 32,000 | tools |
| Claude Opus 4.5 | $5.00/M | $25.00/M | 200,000 | tools · vision · reasoning · cache |
| Claude Opus 4.6 | $5.00/M | $25.00/M | 200,000 | tools · vision · reasoning · cache |
| Claude Opus 4.7 | $5.00/M | $25.00/M | 200,000 | tools · vision · reasoning · cache |
| Meta Llama 3.1 405B Instruct | $5.33/M | $16.00/M | 128,000 | |
| Claude Opus 4.1 | $15.00/M | $75.00/M | 200,000 | tools · vision · reasoning · cache |
| Jais 30B Chat | $3,200/M | $9,710/M | 8,192 |
rerank 5
| Model ↕ | Input / 1M ↕ | Output / 1M ↕ | Context ↕ | Caps |
|---|---|---|---|---|
| Cohere Rerank V3 English | — | — | 4,096 | |
| Cohere Rerank V3 Multilingual | — | — | 4,096 | |
| Cohere Rerank v3.5 | — | — | 4,096 | |
| Cohere Rerank V4.0 Fast | — | — | 32,768 | |
| Cohere Rerank V4.0 Pro | — | — | 32,768 |
ocr 5
| Model ↕ | Input / 1M ↕ | Output / 1M ↕ | Context ↕ | Caps |
|---|---|---|---|---|
| Doc Intelligence Prebuilt Document | — | — | — | |
| Doc Intelligence Prebuilt Layout | — | — | — | |
| Doc Intelligence Prebuilt Read | — | — | — | |
| Mistral Document AI 2505 | — | — | — | |
| Mistral Document AI 2512 | — | — | — |
embedding 3
| Model ↕ | Input / 1M ↕ | Output / 1M ↕ | Context ↕ | Caps |
|---|---|---|---|---|
| Cohere Embed V3 English | $0.1000/M | — | 512 | |
| Cohere Embed V3 Multilingual | $0.1000/M | — | 512 | |
| Embed V 4.0 | $0.120/M | — | 128,000 |
image generation 3
| Model ↕ | Input / 1M ↕ | Output / 1M ↕ | Context ↕ | Caps |
|---|---|---|---|---|
| Flux 1 Kontext Pro | — | — | — | |
| Flux 1.1 Pro | — | — | — | |
| Flux 2 Pro | — | — | — |
FAQ
How many Azure AI Foundry models are there?
76 Azure AI Foundry models are listed across 5 modalities on this page. 71 have public per-token pricing.
How is Azure AI Foundry pricing verified?
Pricing is aggregated from BerriAI/litellm, models.dev, and OpenRouter and refreshed weekly. Each row shows a per-model "verified" date. If a price is wrong, click the row to open the model page and use the inline "suggest edit" link — submissions go into a public review queue.
Which Azure AI Foundry model is cheapest?
Input pricing on Azure AI Foundry starts at $0.0400 per 1M tokens. Sort the table by price (or use the in-page filter at the top) to find the cheapest model that matches your capability requirements.
Can I route to Azure AI Foundry via an OpenAI-compatible API?
Yes — point your OpenAI client at Future AGI's Agent Command Center, configure a Azure AI Foundry target, and call Azure AI Foundry models with the standard /v1/chat/completions surface. The same gateway can route to other providers as fallback. Free for the first 100K requests/month.