LLM API Pricing: Token Prices by Model and Platform

LLM API pricing for text models on Vertex AI, Azure, Gemini API, Mistral and Cohere: input, output, cached and batch prices per 1M tokens, updated weekly.

Last published
Refresh
Weekly
Current rows
990
Licence
Free snapshotCC BY 4.0Fork and collectCommercial
Sample
View on GitHub
The brief
I need this data: LLM API pricing for text models on Vertex AI, Azure, Gemini API, Mistral and Cohere: input, output, cached and batch prices per 1M tokens, updated weekly.

One message. 990 rows, 12 columns.

Latest results

990 rows, refreshed weekly

Row#Open row
12026-10-30Gemini APIGoogleGemini 3.6 Flashstandard0.753.750.0750.3751.875input price: $0.75 through December 31, 2026. $1.5 starting January 1, 2027.; output price (including thinking tokens): $3.75 through December 31, 2026. $7.5 starting January 1, 2027.ai.google.dev/gemini-api/docs/pricingView row
22026-10-30AzureOpenAIGPT-56terra Flglobal160.1——Azure price reference: eastus2; input SKU 56terra ShCo Inp Fl Gl; output SKU 56terra ShCo Opt Fl Gl. Higher-context meters: 56terra LoCo Opt Fl Gl $9/1M; 56terra LoCo Inp Fl Gl $2/1M; 56terra LoCo Cd Inp Fl Gl $0.2/1M.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
32026-10-30AzureOpenAIGPT-5 codexdata zone1.37511———Azure price reference: eastus2; input SKU gpt-5-codex-inp-dzone; output SKU gpt-5-codex-out-dzone.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
42026-10-30AzureMoonshot AIKimi K2.7 Codedata zone1.04550.209——Azure price reference: eastus2; input SKU K2.7 Code Inp DZ; output SKU K2.7 Code Outp DZ.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
52026-10-30AzureOpenAIGPT-4.1 nanoglobal0.10.40.0250.050.2Azure price reference: eastus2; input SKU gpt 4.1 nano Inp glbl; output SKU gpt 4.1 nano Outp glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
62026-10-30AzureOpenAIGPT-4.1 miniglobal0.41.60.10.20.8Azure price reference: eastus2; input SKU gpt 4.1 mini Inp glbl; output SKU gpt 4.1 mini Outp glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
72026-10-30AzureOpenAIGPT-56sol Flglobal2100.2——Azure price reference: eastus2; input SKU 56sol ShCo Inp Fl Gl; output SKU 56sol ShCo Opt Fl Gl. Higher-context meters: 56sol LoCo Opt Fl Gl $15/1M; 56sol LoCo Inp Fl Gl $4/1M; 56sol LoCo Cd Inp Fl Gl $0.4/1M.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
82026-10-30Gemini APIGoogleGemini 2.5 Flashstandard0.32.50.030.151.25—ai.google.dev/gemini-api/docs/pricingView row
92026-10-30AzurexAIGrok 4.1global0.20.5———Azure price reference: eastus2; input SKU Grok 4.1 Inp Glbl; output SKU Grok 4.1 Outp Glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
102026-10-30Gemini APIGoogleGemini 2.5 Prostandard1.25100.1250.6255input price: $1.25, prompts <= 200k tokens $2.5, prompts > 200k tokens; output price (including thinking tokens): $10, prompts <= 200k tokens $15, prompts > 200kai.google.dev/gemini-api/docs/pricingView row
112026-10-30AzureMoonshot AIKimi Model 7global3.30.33———Azure price reference: eastus2; input SKU Model 7 Inp glbl; output SKU Model 7 Outp glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
122026-10-30AzureOpenAIGPT-4 8Kglobal3060———Azure price reference: spaincentral; input SKU gpt-4-8K-Inp-glbl; output SKU gpt-4-8K-Outp-glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
132026-10-30AzureOpenAIGPT-3.5 Turbo Instructregional1.652.2———US East 2 unavailable; fallback regional reference: eastus. Azure price reference: eastus; input SKU gpt-35-turbo-Instruct-Input-regional; output SKU gpt-35-turbo-Instruct-Output-regional.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
142026-10-30AzurexAIGrok Code Fast 1data zone0.221.65———Azure price reference: eastus2; input SKU Code Fast 1 Inp DZone; output SKU Code Fast 1 Outp DZone.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
152026-10-30AzureOpenAIo4-mini 0416regional1.214.840.303——Regional reference: eastus2. Azure price reference: eastus2; input SKU o4-mini 0416 Inp regnl; output SKU o4-mini 0416 Outp regnl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
162026-10-30AzureOpenAIGPT-5.1 codexglobal1.25100.125——Azure price reference: eastus2; input SKU 5.1 codex inp Gl; output SKU 5.1 codex opt Gl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
172026-10-30AzureOpenAIGPT-6 solglobal2100.2——Azure price reference: eastus2; input SKU 6-sol ShortCo Inp Std Gl; output SKU 6-sol ShortCo Opt Std Gl. Higher-context meters: 6-sol LongCo Inp Std Gl $4/1M; 6-sol LongCo Opt Std Gl $15/1M; 6-sol LongCo Cd Inp Std Gl $0.4/1M.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
182026-10-30CohereCohereAya Expanse 32Bstandard0.51.5————cohere.com/pricingView row
192026-10-30AzureOpenAIcomputer useglobal312———Azure price reference: eastus2; input SKU computer-use-inpt-glbl; output SKU computer-use-outp-glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
202026-10-30AzureOpenAIGPT-5 Chatdata zone1.375110.138——Azure price reference: northeurope; input SKU GPT 5 Chat Inpt DZone; output SKU GPT 5 Chat outpt DZone.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
212026-10-30AzureMicrosoftPhi-4 reasoningstandard0.1250.5———Regional reference: eastus2. Azure price reference: eastus2; input SKU Phi-4-reasoning-Input; output SKU Phi-4-reasoning-Output.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
222026-10-30Vertex AIGoogleGemini 3.1 Flash-Literegional0.2751.650.0280.1380.825—cloud.google.com/vertex-ai/generative-ai/pricingView row
232026-10-30AzureMoonshot AIKimi K2.6 Thinkingdata zone1.0455———Azure price reference: eastus2; input SKU K2.6 Thinking Inp DZ; output SKU K2.6 Thinking Outp DZ.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
242026-10-30AzurexAIGrok 4.6data zone2.26.60.55——Azure price reference: eastus2; input SKU 4.6 Inp DZ; output SKU 4.6 Outp DZ. Higher-context meters: 4.6 Cached DZ L $1.1/1M; 4.6 Outp DZ L $13.2/1M; 4.6 Inp DZ L $4.4/1M.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
252026-10-30AzureDeepSeekDeepSeek-V3 0324global1.144.56———Azure price reference: eastus2; input SKU V3-0324 Inp glbl; output SKU V3-0324 Outp glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
262026-10-30Vertex AIGoogleGemini 3.6 Flashglobal0.753.750.0750.3751.875Published price period: Gemini 3.6 Flash * through December 31, 2026.cloud.google.com/vertex-ai/generative-ai/pricingView row
272026-10-30CohereCohereCommand-lightstandard0.30.6———Legacy model price for existing customers, as published in the pricing FAQ.cohere.com/pricingView row
282026-10-30AzureMistralMistral Large 3data zone0.551.65———Azure price reference: eastus2; input SKU Large 3 Inp Dzone; output SKU Large 3 Outp Dzone.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
292026-10-30AzureDeepSeekDeepSeek-V4 Prodata zone1.913.830.16——Azure price reference: eastus2; input SKU V4 Pro Inp DZ; output SKU V4 Pro Outp DZ.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
302026-10-30Vertex AIGoogleGemini 2.5 Flashstandard0.32.50.030.151.25—cloud.google.com/vertex-ai/generative-ai/pricingView row
312026-10-30Vertex AIGoogleGemini 3.5 Flashglobal1.590.15—4.5—cloud.google.com/vertex-ai/generative-ai/pricingView row
322026-10-30AzureOpenAIo1-minidata zone1.214.840.6050.6052.42Azure price reference: eastus2; input SKU o1 mini input Data Zone; output SKU o1 mini output Data Zone.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
332026-10-30AzureOpenAIGPT-6.1 solglobal2100.1——Azure price reference: eastus2; input SKU 6.1-sol ShortCo Inp Std Gl; output SKU 6.1-sol ShortCo Opt Std Gl. Higher-context meters: 6.1-sol LongCo Opt Std Gl $15/1M; 6.1-sol LongCo Inp Std Gl $4/1M; 6.1-sol LongCo Cd Inp Std Gl $0.2/1M.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
342026-10-30AzureOpenAIGPT-56luna Flglobal0.10.60.01——Azure price reference: eastus2; input SKU 56luna ShCo Inp Fl Gl; output SKU 56luna ShCo Opt Fl Gl. Higher-context meters: 56luna LoCo Opt Fl Gl $0.9/1M; 56luna LoCo Cd Inp Fl Gl $0.02/1M; 56luna LoCo Inp Fl Gl $0.2/1M.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
352026-10-30AzureOpenAIGPT-4o 0806global2.5101.251.255Azure price reference: eastus2; input SKU gpt-4o-0806-Inp-glbl; output SKU gpt-4o-0806-Outp-glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
362026-10-30AzureMicrosoftPhi-3 Medium 128K Instructstandard0.170.68———Regional reference: eastus2. Azure price reference: eastus2; input SKU Phi-3-Medium-128K-Instruct-Input; output SKU Phi-3-Medium-128K-Instruct-Output.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
372026-10-30AzurexAIGrok 4.7data zone2.26.60.55——Azure price reference: eastus2; input SKU 4.7 Inp DZ; output SKU 4.7 Outp DZ. Higher-context meters: 4.7 Outp DZ L $13.2/1M; 4.7 Inp DZ L $4.4/1M; 4.7 Cached DZ L $1.1/1M.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
382026-10-30AzureOpenAIGPT-3.5 Turbo 16K 1106standard12———US East 2 unavailable; fallback regional reference: Global. Azure price reference: Global; input SKU Az-GPT35-Turbo-16K-1106 Input; output SKU Az-GPT35-Turbo-16K-1106 Output.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
392026-10-30AzureMistralMistral Large 3global0.51.5———Azure price reference: eastus2; input SKU Large 3 Inp glbl; output SKU Large 3 Outp glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
402026-10-30AzureOpenAIGPT-5.3 chatglobal1.75140.175——Azure price reference: eastus2; input SKU 5.3 chat inp Gl; output SKU 5.3 chat opt Gl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
412026-10-30AzureMiniMaxFW MiniMax 3data zone0.331.320.066——Azure price reference: eastus2; input SKU FW MiniMax 3 Inp DZ; output SKU FW MiniMax 3 Outp DZ.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
422026-10-30AzureOpenAIGPT-4o mini 0718regional0.1650.660.083——Regional reference: eastus2. Azure price reference: eastus2; input SKU gpt-4o-mini-0718-Inp-regnl; output SKU gpt-4o-mini-0718-Outp-regnl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
432026-10-30AzureOpenAIGPT-5 Minidata zone0.2752.20.0280.1381.1Azure price reference: eastus2; input SKU GPT 5 Mini Inpt DZone; output SKU GPT 5 Mini outpt DZone.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
442026-10-30AzureOpenAIFW GPT-OSS-120Bdata zone0.1650.660.082——Azure price reference: eastus2; input SKU FW GPT OSS 120B Inp DZ; output SKU FW GPT OSS 120B Outp DZ.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
452026-10-30AzureOpenAIGPT-5.6 lunadata zone0.221.320.022——Azure price reference: eastus2; input SKU 5.6 luna ShortCo Inp Std DZ; output SKU 5.6 luna ShortCo Opt Std DZ. Higher-context meters: 5.6 luna LongCo Cd Inp Std DZ $0.044/1M; 5.6 luna LongCo Inp Std DZ $0.44/1M; 5.6 luna LongCo Opt Std DZ $1.98/1M.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
462026-10-30AzureOpenAIGPT-3.5 Turbo 16K 1106regional1.12.2———US East 2 unavailable; fallback regional reference: westus. Azure price reference: westus; input SKU gpt-35-turbo-16K-1106 Input-regional; output SKU gpt-35-turbo-16K-1106 Output-regional.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
472026-10-30AzureDeepSeekDeepSeek-V4 Flashglobal0.190.510.028——Azure price reference: eastus2; input SKU V4 Flash Inp glbl; output SKU V4 Flash Outp glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
482026-10-30AzureOpenAIGPT-5.1 codex maxglobal1.25100.125——Azure price reference: eastus2; input SKU 5.1 codex max inp Gl; output SKU 5.1 codex max opt Gl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
492026-10-30AzurexAIGrok 4 Fastglobal0.20.5———Azure price reference: eastus2; input SKU Grok4 Fast Inp glbl; output SKU Grok4 Fast Outp glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
502026-10-30AzureMoonshot AIKimi K2.7 Codeglobal0.9540.19——Azure price reference: eastus2; input SKU K2.7 Code Inp glbl; output SKU K2.7 Code Outp glbl.prices.azure.com/api/retail/prices?$filter=serviceName%20eq%20%27Foundry%20Models%27View row
990 rows × 12 columns

Data sources

cloud.google.com/vertex-ai/generative-ai/pricingprices.azure.com/api/retail/pricesai.google.dev/gemini-api/docs/pricingmistral.ai/pricingcohere.com/pricing

Produced by pipeline v5Operate a site this data was read from? Report a problem or request removal.

About this dataset

LLM API pricing for text models in one table, captured every week from five platforms: Google Vertex AI, Azure, the Gemini API, Mistral and Cohere. Each row is one model on one platform for one deployment type (global, regional, data zone or standard) on one capture date. Prices are in US dollars per 1 million tokens, converted where a platform lists them per 1,000. Where the platform publishes them, rows also carry cached-input and batch prices and a note on long-context pricing.

What you can use it for

Compare what the same model costs on different platforms and deployment types. Estimate the monthly cost of an LLM feature from your expected input and output tokens. See which models offer cached-input or batch discounts. Track list-price changes week by week, since every snapshot is kept.

Coverage

Text-token prices for models from 13 makers, including Google, OpenAI, Anthropic, Meta, Mistral, xAI, DeepSeek and Cohere, as listed on Vertex AI, Azure, the Gemini API, Mistral and Cohere. Snapshots start on 23 Oct 2026, and a new one is added every week. Image, video, audio, live, robotics, OCR and fine-tuning prices are not included.

Sources

Google Cloud Vertex AI generative AI pricing, the Azure Retail Prices API, Google Gemini API pricing, Mistral pricing and Cohere pricing. Each row links to the page or API it came from.

Data pipeline

How Jsonify collects this data

Jason, Jsonify's AI data engineer, built this pipeline from a plain-language brief. It runs the same steps every time: read the sources, extract the fields, check every row, publish the dataset. When a source changes and a step fails, Jason diagnoses the failure, repairs the pipeline and verifies a sample before it resumes, so the dataset keeps its shape without anyone babysitting it.

  1. 01 · input

    History

    Saved weekly price snapshots used to preserve history and…

  2. 02

    Prepare

    Reserve this snapshot within the 5,000-row monthly budget and…

  3. 03 · fan out

    Collect

    Read the approved live pricing pages and all Azure…

  4. 04

    Clean Models

    Exclude video, image and audio generation models, live…

  5. 05

    Validate

    Keep only complete, finite, unique model/deployment quotes…

  6. 06 · output

    Main

    Weekly text-token API prices in USD per million tokens, with…

    990 rows · v5

History

Saved weekly price snapshots used to preserve history and enforce the monthly collection ceiling.

Column explanations

captured_date: date the price snapshot was captured. platform: pricing platform. model_maker: model maker. model: model name. deployment_type: deployment category; values are global, regional, data zone, or standard. input_usd_per_1m: input-token price in USD per million tokens. output_usd_per_1m: output-token price in USD per million tokens. cached_input_usd_per_1m: cached-input price in USD per million tokens, when available. batch_input_usd_per_1m: batch input-token price in USD per million tokens, when available. batch_output_usd_per_1m: batch output-token price in USD per million tokens, when available. long_context_note: long-context note, when available. source_url: URL of the pricing source.

History

Versions, repairs and quality checks

Every activated version of the pipeline, the recorded outcome of each repair, the latest quality checks on published data, and what the dataset does not cover.

Version history

  1. v5

    Corrected Gemini billing-tier separation and Azure abbreviated token-meter labels to restore missing deployment offers and cache prices.

  2. v4

    Exclude Gemini Omni and other video, image and audio generation models from future text-token price snapshots, while retaining Computer Use-Preview.

  3. v3

    Fixed saved price history loading so updates can proceed while retaining prior snapshots and duplicate protection.

Delivery options

Using this data

Licence

Use this snapshot, free

Jsonify collected this snapshot and publishes it under CC BY 4.0. Use it for anything, including commercially, as long as you credit Jsonify with a link to this page.

Credit it asData: Jsonify, CC BY 4.0, https://www.jsonify.com/datasets/llm-api-pricing-30a0181a4f3a

The licence covers the dataset as Jsonify compiled it. It grants no rights in the source websites’ own content or trademarks.

Or collect your own

Fork this pipeline into your own Jsonify account and run it yourself, or have us run a live feed for your team.

  • Commercial use, no credit line. Data you collect is yours under your Jsonify terms.
  • Kept current. It runs on your schedule, and Jason repairs it when a source changes.
  • Change alerts. Hear about new and changed rows in Slack, Teams or email.
  • Delivered where you work. Google Sheets, Excel, Snowflake, Postgres, S3 and more.
  • Your changes. Different sources, columns or coverage: tell Jason what to change.

100 free rows a month, then pay per row. See pricing · Book a demo for a team feed

Frequently asked questions

Is this dataset free to use?
Yes. The snapshot on this page is published under CC BY 4.0. You can use it in your own work, products and research, including commercially, as long as you credit Jsonify.
How do I credit Jsonify?
Put this line wherever you show or publish the data, and say if you changed it: Data: Jsonify, CC BY 4.0, https://www.jsonify.com/datasets/llm-api-pricing-30a0181a4f3a
How is that different from collecting my own?
The free snapshot is the data Jsonify collected, as of the date shown on this page. If you run the pipeline in your own Jsonify account, or have us run a live feed for you, the data you collect is yours to use under your Jsonify terms. It needs no credit line, and it stays current on your schedule.
What do Download and Customize do?
Both open a free Jsonify workspace of your own. Download puts this dataset in it, ready to export as CSV or JSON. Customize puts a copy of the pipeline in it, so you can change the sources, the rows it collects or the schedule.
Does collecting my own cost anything?
Your account includes free rows to start. Past that you pay for the rows you collect, as set out on the pricing page. Nothing is charged for the Download or Customize step itself.
What does a live feed for my team include?
Jsonify runs the pipeline for you on a schedule, sends change alerts, keeps the history, and adds the coverage you need. It is priced per team. Book a demo to scope it.

Need something different?

Describe it. Get verified rows, kept current.

Build it →Book a demo
Competitor prices Restaurant menus Job listings News monitoring Property listings Customer reviews Grocery prices Insurance quotes Rental rates Broadband offers

Migrate from your current tool

Share what you run on your current tool today. Jason, Jsonify’s AI data engineer, reads it and proposes the same pipelines on Jsonify, with checks and repairs built in.

Drop files or browseExports, task or actor configs, sample datasets, screenshots. Up to 20 files, 50 MB each.

    Don’t include passwords or API keys. Files are shared with Jason to plan the migration. Privacy

    Connect your data assistant

    Build datasets and work with your data in Claude, ChatGPT, Copilot or another assistant.

    Connect in Claude

    1. Open the Jsonify connector: Open on Claude.ai ↗
    2. Select Connect and sign in to Jsonify.
    3. In a chat, enable Jsonify under + → Connectors and describe the data you need.
    Advanced

    Team or Enterprise account? Your workspace owner enables Jsonify once for everyone from Admin settings → Connectors. Jsonify is a community connector in the directory — some admins allow those by default, some review them first.

    No directory? Add it by hand: Settings → Connectors → + Add custom connector, name it Jsonify, paste the server URL below, then Connect.

    Server URLhttps://factory.jsonify.com/mcp

    Official Claude custom-connector guide ↗

    Then say: “build me a dataset of competitor product prices and availability, refreshed daily”. Full instructions per client on /connect.