BroadRouter

Azure OpenAI and Foundry Models API pricing

Updated

Azure OpenAI and Foundry Models lists standard prices for 97 models. The cheapest is NVIDIA Nemotron 3.5 Lightning 30B A3B at USD 0.06 input and USD 0.22 output per 1M tokens; the dearest is GPT-5.4 Pro at USD 30 / USD 180. It publishes cached-input prices for 59 of them, plus batch, flex / spot, priority and fast tiers.

Azure OpenAI and Foundry Models hosts models it does not train. By maker: OpenAI 44, Microsoft 17, xAI 10, DeepSeek 9, Moonshot AI 5, Z.ai 3, and 6 more. 124 prices, each read from Azure OpenAI and Foundry Models' own pricing page or one of 8 linked model pages, checked 2026-11-14. BroadRouter is BroadNet's own site and does not sell Azure OpenAI and Foundry Models' services. Automated re-check 2026-11-16: 124 of 124 prices still matched the provider's page.

Models with a standard price
97
Median input / output
USD 1.10 / USD 4.40
Cached-input prices
59 of 97
Cheapest host for
0 of 53 shared models

Which models is Azure OpenAI and Foundry Models the cheapest provider for?

None outright. Of the 53 models that Azure OpenAI and Foundry Models shares with other providers, 24 are cheaper elsewhere at the standard tier, and Azure OpenAI and Foundry Models ties the lowest price for the other 29, where another provider charges the same. It is the dearest provider for 14.

The 6 models Azure OpenAI and Foundry Models sells that the most other providers also sell, with Azure OpenAI and Foundry Models' rank among them on the 3:1 blend.

Azure OpenAI and Foundry Models on the most widely hosted models
ModelAzure OpenAI and Foundry ModelsRankCheapest
OpenAI GPT OSS 120BUSD 0.15 / USD 0.6011 of 21CoreWeave Forge USD 0.03 / USD 0.17
GLM 5.3 FlashUSD 0.188 / USD 0.62520 of 20GMI Cloud USD 0.075 / USD 0.25
GLM 5.2USD 1.54 / USD 4.8419 of 19DeepInfra USD 0.5625 / USD 1.80
Kimi K3USD 3 / USD 155 of 18Chutes USD 1.85 / USD 9.50
DeepSeek V4.1 FlashUSD 0.30 / USD 1.205 of 16DeepInfra USD 0.14 / USD 0.42
DeepSeek V4 Flash 0731USD 0.44 / USD 1.3210 of 15DeepInfra USD 0.06 / USD 0.18

Which model makers does Azure OpenAI and Foundry Models sell?

Azure OpenAI and Foundry Models prices models from 12 makers. OpenAI leads with 44 models. Standard prices, USD per 1M input / output tokens.

Azure OpenAI and Foundry Models models by maker
MakerModelsCheapest modelMedian input / output
OpenAI44GPT-5 Nano USD 0.05 / USD 0.40USD 1.50 / USD 10
Microsoft17Phi-4-mini USD 0.075 / USD 0.30USD 0.13 / USD 0.52
xAI10Grok 4 Fast USD 0.20 / USD 0.50USD 1.25 / USD 2.50
DeepSeek9DeepSeek V4 Flash USD 0.19 / USD 0.51USD 1.14 / USD 3.48
Moonshot AI5Kimi K2 Thinking USD 0.60 / USD 2.50USD 0.95 / USD 4
Z.ai3GLM 5.3 Flash USD 0.188 / USD 0.625USD 1.54 / USD 4.84
Cohere2Command A+ USD 0.80 / USD 3.20USD 1.65 / USD 6.60
Meta2Llama 4 Maverick 17B 128E Instruct FP8 USD 0.25 / USD 1USD 0.48 / USD 0.855
NVIDIA2NVIDIA Nemotron 3.5 Lightning 30B A3B USD 0.06 / USD 0.22USD 0.36 / USD 1.43
MiniMax1MiniMax M3 USD 0.33 / USD 1.32—
Mistral AI1Mistral Medium 3.5 USD 1.50 / USD 7.50—
Thinking Machines1FW Inkling USD 1.10 / USD 4.46—

Azure OpenAI and Foundry Models price list

USD per 1M tokens: 124 prices for 97 models, one row per model with any other service tiers stacked under the standard price. Sources checked 2026-11-14 unless a row says otherwise. "vs maker" compares a standard price with the model maker's own API on the 3:1 blend; a minus means Azure OpenAI and Foundry Models is cheaper. Of 38 hosted model prices with a maker price to compare, 0 are below the maker's and 6 above.

Azure OpenAI and Foundry Models API prices by model
ModelTierInputCachedOutputContextvs makerSource
Cohere Command AstandardUSD 2.50—USD 10not stated—source
Command A+standardUSD 0.80—USD 3.20not stated—source
DeepSeek R1standardUSD 1.35—USD 5.40not stated—source
DeepSeek V3standardUSD 1.14—USD 4.56not stated—source
DeepSeek V3 0324standardUSD 1.14—USD 4.56not stated—source
DeepSeek V3.1standardUSD 1.23—USD 4.94not stated—source
DeepSeek V3.2standardUSD 0.58—USD 1.68not stated—source
DeepSeek V4 FlashstandardUSD 0.19USD 0.028USD 0.51not stated—source
DeepSeek V4 Flash 0731standardUSD 0.44USD 0.014USD 1.32not stated—source
DeepSeek V4 ProstandardUSD 1.74USD 0.145USD 3.48not stated+10%source
DeepSeek V4.1 FlashstandardUSD 0.30USD 0.006USD 1.20not statedsamesource
FW InklingstandardUSD 1.10USD 0.19USD 4.46not stated—source
GLM 5.1standardUSD 1.54USD 0.286USD 4.84not stated+10%source
GLM 5.2standard
fast
USD 1.54
USD 2.31
USD 0.15
USD 0.231
USD 4.84
USD 7.26
not stated+10%source
GLM 5.3 FlashstandardUSD 0.188USD 0.038USD 0.625not stated+25%source
GPT Chat LateststandardUSD 5USD 0.50USD 30not statedsamesource
GPT-4.1 (2025-04-14)standard
batch
priority
USD 2
USD 1
USD 3.50
USD 0.50
—
USD 0.88
USD 8
USD 4
USD 14
not statedsamesource
GPT-4.1 Ministandard
batch
priority
USD 0.40
USD 0.20
USD 0.70
USD 0.10
—
USD 0.18
USD 1.60
USD 0.80
USD 2.80
not statedsamesource
GPT-4.1 nano (2025-04-14)standard
batch
USD 0.10
USD 0.05
USD 0.03
—
USD 0.40
USD 0.20
not stated—source
GPT-4o (2024-05-13)standard
batch
USD 5
USD 2.50
—
—
USD 15
USD 7.50
not stated—source
GPT-4o (2024-08-06)standard
batch
USD 2.50
USD 1.25
USD 1.25
—
USD 10
USD 5
not stated—source
GPT-4o (2024-11-20)standard
batch
USD 2.50
USD 1.25
USD 1.25
—
USD 10
USD 5
not stated—source
GPT-4o ministandard
batch
USD 0.15
USD 0.075
USD 0.075
—
USD 0.60
USD 0.30
not statedsamesource
GPT-5 (2025-08-07)standard
batch
USD 1.25
USD 0.63
USD 0.13
USD 0.07
USD 10
USD 5
not statedsamesource
GPT-5 ChatstandardUSD 1.25USD 0.13USD 10not stated—source
GPT-5 CodexstandardUSD 1.25USD 0.13USD 10not stated—source
GPT-5 MinistandardUSD 0.25USD 0.03USD 2not statedsamesource
GPT-5 NanostandardUSD 0.05USD 0.01USD 0.40not statedsamesource
GPT-5 ProstandardUSD 15—USD 120not statedsamesource
GPT-5.1standard
priority
USD 1.25
USD 2.50
USD 0.13
USD 0.25
USD 10
USD 20
not statedsamesource
GPT-5.1 ChatstandardUSD 1.25USD 0.13USD 10not stated—source
GPT-5.1 CodexstandardUSD 1.25USD 0.13USD 10not stated—source
GPT-5.1 Codex MaxstandardUSD 1.25USD 0.13USD 10not stated—source
GPT-5.1 Codex MinistandardUSD 0.25USD 0.03USD 2not stated—source
GPT-5.2standard
priority
USD 1.75
USD 3.50
USD 0.18
USD 0.35
USD 14
USD 28
not statedsamesource
GPT-5.2 Chat (latest)standardUSD 1.75USD 0.18USD 14not stated—source
GPT-5.2 CodexstandardUSD 1.75USD 0.18USD 14not stated—source
GPT-5.3 ChatstandardUSD 1.75USD 0.18USD 14not stated—source
GPT-5.3 CodexstandardUSD 1.75USD 0.18USD 14not statedsamesource
GPT-5.4standard
batch
priority
USD 2.50
USD 1.25
USD 5
USD 0.25
USD 0.13
USD 0.50
USD 15
USD 7.50
USD 30
not statedsamesource
GPT-5.4 Ministandard
batch
priority
USD 0.75
USD 0.38
USD 1.50
USD 0.08
USD 0.04
USD 0.15
USD 4.50
USD 2.25
USD 9
not statedsamesource
GPT-5.4 NanostandardUSD 0.20USD 0.02USD 1.25not statedsamesource
GPT-5.4 ProstandardUSD 30—USD 180not statedsamesource
GPT-5.5standard
priority
USD 5
USD 12.50
USD 0.50
USD 1.25
USD 30
USD 75
not statedsamesource
GPT-5.6 Lunastandard
flex / spot
USD 0.20
USD 0.10
USD 0.02
USD 0.01
USD 1.20
USD 0.60
not statedsamesource
GPT-5.6 Solstandard
flex / spot
priority
USD 4
USD 2
USD 8
USD 0.40
USD 0.20
USD 0.80
USD 20
USD 10
USD 40
not statedsamesource
GPT-5.6 Terrastandard
flex / spot
priority
USD 2
USD 1
USD 4
USD 0.20
USD 0.10
USD 0.40
USD 12
USD 6
USD 24
not statedsamesource
GPT-6 AstrastandardUSD 10USD 1USD 50not statedsamesource
GPT-6 LunastandardUSD 0.10USD 0.01USD 0.50not statedsamesource
GPT-6 Solstandard
priority
USD 2
USD 4
USD 0.20
USD 0.40
USD 10
USD 20
not statedsamesource
GPT-6.1 SolstandardUSD 2USD 0.10USD 10not statedsamesource
GPT-6.1 TerrastandardUSD 1USD 0.05USD 5not statedsamesource
Grok 4standardUSD 3—USD 15not stated—source
Grok 4 FaststandardUSD 0.20—USD 0.50not stated—source
Grok 4.1 FaststandardUSD 0.20—USD 0.50not stated—source
Grok 4.20 (reasoning)standardUSD 1.25—USD 2.50not statedsamesource
Grok 4.3standardUSD 1.25—USD 2.50not statedsamesource
Grok 4.6standardUSD 2USD 0.50USD 6not statedsamesource
Grok 4.7standardUSD 2USD 0.50USD 6not statedsamesource
Grok Code Fast 1standardUSD 0.20—USD 1.50not stated—source
Grok-3standardUSD 3—USD 15not stated—source
Grok-3 MinistandardUSD 0.25—USD 1.27not stated—source
Kimi K2 ThinkingstandardUSD 0.60—USD 2.50not stated—source
Kimi K2.5standardUSD 0.60USD 0.10USD 3not stated—source
Kimi K2.6standardUSD 1.045USD 0.176USD 4.40not stated+10%source
Kimi K2.7 CodestandardUSD 0.95USD 0.19USD 4not statedsamesource
Kimi K3standardUSD 3USD 0.30USD 15not statedsamesource
Llama 3.3 70BstandardUSD 0.71—USD 0.71not stated—source
Llama 4 Maverick 17B 128E Instruct FP8standardUSD 0.25—USD 1not stated—source
MAI-Cyber-1 FlashstandardUSD 0.60USD 0.06USD 3.50not stated—source
MAI-DS-R1standardUSD 1.35—USD 5.40not stated—source
MAI-Thinking-1standardUSD 2USD 0.20USD 8not stated—source
MiniMax M3standardUSD 0.33USD 0.066USD 1.32not stated+10%source
Mistral Medium 3.5standardUSD 1.50—USD 7.50not stated—source
NVIDIA Nemotron 3 UltrastandardUSD 0.66USD 0.13USD 2.64not stated—source
NVIDIA Nemotron 3.5 Lightning 30B A3BstandardUSD 0.06USD 0.01USD 0.22not stated—source
o1 (2024-12-17)standardUSD 15USD 7.50USD 60not stated—source
o1-mini (2024-09-12)standardUSD 1.10USD 0.55USD 4.40not stated—source
o3 (2025-04-16)standard
batch
USD 2
USD 1
USD 0.50
—
USD 8
USD 4
not statedsamesource
o3 Deep ResearchstandardUSD 10USD 2.50USD 40not stated—source
o3-mini (2025-01-31)standard
batch
USD 1.10
USD 0.55
USD 0.55
—
USD 4.40
USD 2.20
not stated—source
o4-mini (2025-04-16)standard
batch
USD 1.10
USD 0.55
USD 0.28
—
USD 4.40
USD 2.20
not stated—source
OpenAI GPT OSS 120BstandardUSD 0.15—USD 0.60not stated—source
Phi-3-medium-128k-instructstandardUSD 0.17—USD 0.68131,072—source
Phi-3-medium-4k-instructstandardUSD 0.17—USD 0.684,096—source
Phi-3-mini-128k-instructstandardUSD 0.13—USD 0.52131,072—source
Phi-3-mini-4k-instructstandardUSD 0.13—USD 0.524,096—source
Phi-3-small-128k-instructstandardUSD 0.15—USD 0.60131,072—source
Phi-3-small-8k-instructstandardUSD 0.15—USD 0.608,192—source
Phi-3.5-mini-instructstandardUSD 0.13—USD 0.52131,072—source
Phi-3.5-MoE-instructstandardUSD 0.16—USD 0.64131,072—source
Phi-4standardUSD 0.125—USD 0.50131,072—source
Phi-4-ministandardUSD 0.075—USD 0.30131,072—source
Phi-4-mini-reasoningstandardUSD 0.075—USD 0.30131,072—source
Phi-4-multimodal, text and imagestandardUSD 0.08—USD 0.32131,072—source
Phi-4-reasoningstandardUSD 0.125—USD 0.5032,768—source
Phi-4-reasoning-plusstandardUSD 0.125—USD 0.5032,768—source
  • Cached "—" means Azure OpenAI and Foundry Models' page shows no cache price for that row; budget cached tokens at the input price.
  • Context "not stated": the source does not give it (14 of 124 rows have one).

Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).

What service tiers does Azure OpenAI and Foundry Models publish?

  • batch (13 rows). Across 13 models with both prices, the median batch price is 50% below standard. Azure OpenAI and Foundry Models row note: “Azure OpenAI Global deployment, "Pricing with Batch API (1M Tokens)" column.”
  • flex / spot (3 rows). Across 3 models with both prices, the median flex / spot price is 50% below standard. Azure OpenAI and Foundry Models row note: “Azure OpenAI Global deployment, "Flex Pricing (1M Tokens)" column. Short-context rate (≤ 272K input tokens); long-context rows are higher. Cache writes $2.5 per 1M.”
  • priority (10 rows). Across 10 models with both prices, the median priority price is 2× standard (range 1.75× standard to 2.5× standard). Azure OpenAI and Foundry Models row note: “Azure OpenAI Global deployment, "Priority Processing (1M Tokens)" column.”
  • fast (1 row). Across 1 model with both prices, the median fast price is 1.5× standard. Azure OpenAI and Foundry Models row note: “Foundry Models serverless, served by Fireworks on Azure, Data Zone deployment.”

Cached input: 59 of 97 standard prices include a cache-read price, a median 10% of the input price. What cached input means.

What we recorded about Azure OpenAI and Foundry Models' pricing page: Azure OpenAI: Global (global standard) deployment rows, short-context rates where a model has long-context pricing; Flex, Priority and Batch columns stored as separate tiers where published. Data Zone and Regional rows (about 10% higher) are not stored. Foundry Models (serverless / models-as-a-service): Global rows where listed; rows with no deployment label use the East US value from the page data; Fireworks-served models are often Data Zone only (noted per row). Prices in USD. Claude on Azure has no Azure pricing page (sold via Marketplace) and is left out. Image, embedding, rerank, OCR, audio and realtime models are left out. Legacy regional-only GPT-3.5/GPT-4 rows are left out.

Is Azure OpenAI and Foundry Models cheaper than the model makers' own APIs?

Mostly it charges the same as the maker. Of 38 standard prices for models whose maker sells its own API, 0 are below the maker's price, 6 above and 32 the same, on the 3:1 blend. The median difference is zero. The biggest markup is GLM 5.3 Flash, 25% above Z.ai.

BroadNet Inference and Azure OpenAI and Foundry Models

BroadRouter is BroadNet's own site, so BroadNet is never ranked on this page. BroadNet Inference also serves 1 model that Azure OpenAI and Foundry Models prices: DeepSeek V4.1 Flash at USD 0.175 / USD 0.52 against Azure OpenAI and Foundry Models' USD 0.30 / USD 1.20 (BroadNet is cheaper on the 3:1 blend). BroadNet prices are from its rate card, synced 2026-11-14. BroadNet Inference pricing.

More on Azure OpenAI and Foundry Models

Questions

How much does the Azure OpenAI and Foundry Models API cost per token?

Azure OpenAI and Foundry Models' standard prices run from USD 0.06 input and USD 0.22 output per 1M tokens (NVIDIA Nemotron 3.5 Lightning 30B A3B) to USD 30 / USD 180 (GPT-5.4 Pro). The median across 97 standard prices is USD 1.10 input and USD 4.40 output per 1M, which is USD 0.0000011 per input token. Input, cached input and output are billed separately.

What is the cheapest model on Azure OpenAI and Foundry Models?

NVIDIA Nemotron 3.5 Lightning 30B A3B, at USD 0.06 input and USD 0.22 output per 1M tokens (USD 0.10 on a 3:1 input-to-output blend). The cheapest from other makers: GPT-5 Nano (OpenAI) at USD 0.05 / USD 0.40; Phi-4-mini (Microsoft) at USD 0.075 / USD 0.30; Grok 4 Fast (xAI) at USD 0.20 / USD 0.50.

Is Azure OpenAI and Foundry Models cheaper than other providers?

It depends on the model. Of 53 models that other providers also price, Azure OpenAI and Foundry Models is the cheapest standard price for none; it ties the lowest price, without beating it, for 29; another provider is cheaper for 24; and it is the dearest for 14. Its median rank is 1 of a median 6 providers, on a 3:1 input-to-output blend.

Does Azure OpenAI and Foundry Models have batch, cached or priority pricing?

Cached input: yes, for 59 of 97 standard prices, at a median 10% of the input price. Other tiers: batch on 13 rows (median input 50% below standard); flex / spot on 3 rows (median input 50% below standard); priority on 10 rows (median input 2× standard); fast on 1 row (median input 1.5× standard).

How much does a chatbot cost per month on Azure OpenAI and Foundry Models?

For 2,000 requests a day, 1,500 input tokens (60% cached), 300 output, over 30 days: USD 6.66 on NVIDIA Nemotron 3.5 Lightning 30B A3B, USD 133.92 on o4-mini (2025-04-16) (the middle of Azure OpenAI and Foundry Models' list) and USD 5,940 on GPT-5.4 Pro. Try your own numbers.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity