Skip to content

Free AI tool · dated public prices

LLM API pricing, compared and calculated.

Current OpenAI, Anthropic Claude, Google Gemini, Mistral, DeepSeek, xAI, Meta, Cohere and hosted Llama models in one sortable table, priced per 1M input, cached and output tokens. Enter your requests and token counts to see the cost per day, month and year, side by side. Each price shows the date it was checked, and every change is logged.

Your usage

Billed as output. Leave at 0 for non-reasoning use.

Share of each prompt read from the provider's prompt cache.

Batch API

Asynchronous, cheaper where the provider offers it.

Side by side

Tick up to 4 models in the table. Prices checked on 18 October 2026.

GPT-6.1 Sol
OpenAI · gpt-6.1-sol
Per month
$273.75
$9.00 a day · $3,285.00 a year
$0.0090 per request
Claude Haiku 5.5
Anthropic · claude-haiku-5-5
Per month
$13.69
$0.450 a day · $164.25 a year
$0.0005 per request
Gemini 4 Argon
Google · gemini-4-argon
Per month
$273.75
$9.00 a day · $3,285.00 a year
$0.0090 per request
Providers

91 models shown. Cheapest for this usage: Llama 3.1 8B at $4.26 a month.

LLM API prices per 1M tokens and your cost per request, day, month and year. Prices checked on 18 October 2026.
CompareBatch in / outPer dayPer year
Llama 3.1 8Bllama-3.1-8b-instantGroq (Llama)$0.05-$0.08-131K$0.0001$0.140$4.26$51.10
Command R7Bcommand-r7b-12-2024Cohere$0.0375-$0.15-128K$0.0002$0.150$4.56$54.75
Command R7B Arabiccommand-r7b-arabic-02-2025Cohere$0.0375-$0.15-128K$0.0002$0.150$4.56$54.75
Ministral 3 3Bministral-3b-2512Mistral$0.10$0.01$0.10-131K$0.0003$0.250$7.60$91.25
GPT-5 Nanogpt-5-nanoOpenAI$0.05$0.005$0.40$0.025 / $0.20400K$0.0003$0.300$9.13$109.50
Muse Spark 1.2 Contributormuse-spark-1.2-contributorMeta$0.10$0.002$0.20-1.05M$0.0003$0.300$9.13$109.50
Muse Spark 1.3 Contributormuse-spark-1.3-contributorMeta$0.10$0.002$0.20-1.05M$0.0003$0.300$9.13$109.50
Ministral 3 8Bministral-8b-2512Mistral$0.15$0.015$0.15-262K$0.0004$0.375$11.41$136.87
Gemini 2.5 Flash-Litegemini-2.5-flash-liteGoogle$0.10$0.01$0.40$0.05 / $0.201.05M$0.0004$0.400$12.17$146.00
Claude Haiku 5.5claude-haiku-5-5Anthropic$0.10$0.01$0.50$0.05 / $0.251M$0.0005$0.450$13.69$164.25
GPT-6 Lunagpt-6-lunaOpenAI$0.10$0.01$0.50$0.05 / $0.251.05M$0.0005$0.450$13.69$164.25
Ministral 3 14Bministral-14b-2512Mistral$0.20$0.02$0.20-262K$0.0005$0.500$15.21$182.50
Command Rcommand-r-08-2024Cohere$0.15-$0.60-128K$0.0006$0.600$18.25$219.00
DeepSeek V4.1 Flash 1017deepseek-flashDeepSeek$0.15$0.003$0.60-1M$0.0006$0.600$18.25$219.00
GPT-4o minigpt-4o-miniOpenAI$0.15$0.075$0.60$0.075 / $0.30128K$0.0006$0.600$18.25$219.00
Mistral Small 4mistral-small-2603Mistral$0.15$0.015$0.60-262K$0.0006$0.600$18.25$219.00
GPT-5.6 Lunagpt-5.6-lunaOpenAI$0.20$0.02$1.20$0.10 / $0.601.05M$0.0010$1.00$30.42$365.00
GPT-5.4 nanogpt-5.4-nanoOpenAI$0.20$0.02$1.25$0.10 / $0.63400K$0.0010$1.03$31.18$374.13
Codestral 25.08codestral-2508Mistral$0.30$0.03$0.90-256K$0.0011$1.05$31.94$383.25
Gemini 3.1 Flash Litegemini-3.1-flash-liteGoogle$0.25$0.025$1.50$0.13 / $0.751.05M$0.0013$1.25$38.02$456.25
GPT-5 Minigpt-5-miniOpenAI$0.25$0.025$2.00$0.13 / $1.00400K$0.0015$1.50$45.63$547.50
Llama 3.3 70Bllama-3.3-70b-versatileGroq (Llama)$0.59-$0.79-131K$0.0016$1.58$47.91$574.88
GPT-4.1 minigpt-4.1-miniOpenAI$0.40$0.10$1.60$0.20 / $0.801.05M$0.0016$1.60$48.67$584.00
Mistral Large 3mistral-large-2512Mistral$0.50$0.05$1.50-262K$0.0018$1.75$53.23$638.75
Gemini 2.5 Flashgemini-2.5-flashGoogle$0.30$0.03$2.50$0.15 / $1.251.05M$0.0019$1.85$56.27$675.25
Gemini 3.5 Flash Litegemini-3.5-flash-liteGoogle$0.30$0.03$2.50$0.15 / $1.251.05M$0.0019$1.85$56.27$675.25
DeepSeek V4 Prodeepseek-v4-proDeepSeek$0.66$0.022$1.98-1M$0.0023$2.31$70.26$843.15
Mistral Large 4mistral-large-4 · betaMistral$0.68$0.07$2.09-1.05M$0.0024$2.41$73.15$877.82
Gemini 3 Flash Previewgemini-3-flash-previewGoogle$0.50$0.05$3.00$0.25 / $1.501.05M$0.0025$2.50$76.04$912.50
Llama 3.3 70Bmeta-llama/Llama-3.3-70B-Instruct-TurboTogether AI (Llama)$1.04-$1.04-131K$0.0026$2.60$79.08$949.00
Grok Build 0.1grok-build-0.1xAI$1.00$0.20$2.00-256K$0.0030$3.00$91.25$1,095.00
Gemini 3.6 Flashgemini-3.6-flashGoogle$0.75$0.075$3.75$0.38 / $1.881.05M$0.0034$3.38$102.66$1,231.88
Gemini 3.7 Flashgemini-3.7-flashGoogle$0.75$0.075$3.75$0.38 / $1.881.05M$0.0034$3.38$102.66$1,231.88
Gemini 3.8 Flashgemini-3.8-flashGoogle$0.75$0.075$3.75$0.38 / $1.881.05M$0.0034$3.38$102.66$1,231.88
GPT-5.4 minigpt-5.4-miniOpenAI$0.75$0.075$4.50$0.38 / $2.25400K$0.0038$3.75$114.06$1,368.75
Grok 4.20 (Non-Reasoning)grok-4.20-0309-non-reasoningxAI$1.25$0.20$2.50$1.00 / $2.001M$0.0038$3.75$114.06$1,368.75
Grok 4.20 (Reasoning)grok-4.20-0309-reasoningxAI$1.25$0.20$2.50$1.00 / $2.001M$0.0038$3.75$114.06$1,368.75
Grok 4.20 Multi-Agentgrok-4.20-multi-agent-0309xAI$1.25$0.20$2.50$1.00 / $2.001M$0.0038$3.75$114.06$1,368.75
Grok 4.3grok-4.3xAI$1.25$0.20$2.50$1.00 / $2.001M$0.0038$3.75$114.06$1,368.75
Claude Haiku 4.5claude-haiku-4-5Anthropic$1.00$0.10$5.00$0.50 / $2.50200K$0.0045$4.50$136.88$1,642.50
Muse Spark 1.1muse-spark-1.1Meta$1.25$0.15$4.25-1.05M$0.0046$4.63$140.68$1,688.13
Muse Spark 1.2muse-spark-1.2Meta$1.25$0.15$4.25-1.05M$0.0046$4.63$140.68$1,688.13
Muse Spark 1.3muse-spark-1.3Meta$1.25$0.15$4.25-1.05M$0.0046$4.63$140.68$1,688.13
Mistral Medium 3.5mistral-medium-2604Mistral$1.50$0.15$7.50-262K$0.0068$6.75$205.31$2,463.75
Grok 4.5grok-4.5xAI$2.00$0.30$6.00-500K$0.0070$7.00$212.92$2,555.00
Grok 4.6grok-4.6xAI$2.00$0.50$6.00-500K$0.0070$7.00$212.92$2,555.00
Grok 4.7grok-4.7xAI$2.00$0.50$6.00-500K$0.0070$7.00$212.92$2,555.00
Gemini 2.5 Progemini-2.5-proGoogle$1.25$0.13$10.00$0.63 / $5.001.05M$0.0075$7.50$228.13$2,737.50
Gemini 3.5 Flashgemini-3.5-flashGoogle$1.50$0.15$9.00$0.75 / $4.501.05M$0.0075$7.50$228.13$2,737.50
GPT-5gpt-5OpenAI$1.25$0.13$10.00$0.63 / $5.00400K$0.0075$7.50$228.13$2,737.50
GPT-5.1gpt-5.1OpenAI$1.25$0.13$10.00$0.63 / $5.00400K$0.0075$7.50$228.13$2,737.50
GPT-4.1gpt-4.1OpenAI$2.00$0.50$8.00$1.00 / $4.001.05M$0.0080$8.00$243.33$2,920.00
o3o3OpenAI$2.00$0.50$8.00$1.00 / $4.00200K$0.0080$8.00$243.33$2,920.00
Claude Sonnet 5claude-sonnet-5Anthropic$2.00$0.20$10.00$1.00 / $5.001M$0.0090$9.00$273.75$3,285.00
Claude Sonnet 5.5claude-sonnet-5-5Anthropic$2.00$0.10$10.00$1.00 / $5.001M$0.0090$9.00$273.75$3,285.00
Gemini 4 Argongemini-4-argonGoogle$2.00$0.10$10.00$1.00 / $5.001.05M$0.0090$9.00$273.75$3,285.00
GPT-6 Solgpt-6-solOpenAI$2.00$0.20$10.00$1.00 / $5.001.05M$0.0090$9.00$273.75$3,285.00
GPT-6.1 Solgpt-6.1-solOpenAI$2.00$0.10$10.00$1.00 / $5.001.05M$0.0090$9.00$273.75$3,285.00
Command Acommand-a-03-2025Cohere$2.50-$10.00-256K$0.010$10.00$304.17$3,650.00
Command A Pluscommand-a-plus-05-2026Cohere$2.50-$10.00-128K$0.010$10.00$304.17$3,650.00
Command A Reasoningcommand-a-reasoning-08-2025Cohere$2.50-$10.00-256K$0.010$10.00$304.17$3,650.00
Command A Visioncommand-a-vision-07-2025Cohere$2.50-$10.00-128K$0.010$10.00$304.17$3,650.00
Command R+command-r-plus-08-2024Cohere$2.50-$10.00-128K$0.010$10.00$304.17$3,650.00
Gemini 3.1 Pro Previewgemini-3.1-pro-previewGoogle$2.00$0.20$12.00$1.00 / $6.001.05M$0.010$10.00$304.17$3,650.00
GPT-4ogpt-4oOpenAI$2.50$1.25$10.00$1.25 / $5.00128K$0.010$10.00$304.17$3,650.00
GPT-5.6 Terragpt-5.6-terraOpenAI$2.00$0.20$12.00$1.00 / $6.001.05M$0.010$10.00$304.17$3,650.00
GPT-5.2gpt-5.2OpenAI$1.75$0.18$14.00$0.88 / $7.00400K$0.011$10.50$319.38$3,832.50
GPT-5.3 Codexgpt-5.3-codexOpenAI$1.75$0.18$14.00-400K$0.011$10.50$319.38$3,832.50
GPT-5.3 Codex Sparkgpt-5.3-codex-sparkOpenAI$1.75$0.18$14.00-128K$0.011$10.50$319.38$3,832.50
GPT-5.4gpt-5.4OpenAI$2.50$0.25$15.00$1.25 / $7.501.05M$0.013$12.50$380.21$4,562.50
Claude Sonnet 4.5claude-sonnet-4-5Anthropic$3.00$0.30$15.00$1.50 / $7.50200K$0.014$13.50$410.63$4,927.50
Claude Sonnet 4.6claude-sonnet-4-6Anthropic$3.00$0.30$15.00$1.50 / $7.501M$0.014$13.50$410.63$4,927.50
Claude Opus 5.5claude-opus-5-5Anthropic$4.00$0.20$20.00$2.00 / $10.001M$0.018$18.00$547.50$6,570.00
GPT-5.6gpt-5.6OpenAI$4.00$0.40$20.00$2.00 / $10.001.05M$0.018$18.00$547.50$6,570.00
GPT-5.6 Solgpt-5.6-solOpenAI$4.00$0.40$20.00$2.00 / $10.001.05M$0.018$18.00$547.50$6,570.00
Muse Wondermuse-wonder-1.0Meta$4.00$0.20$20.00-1.05M$0.018$18.00$547.50$6,570.00
Claude Opus 4.5claude-opus-4-5Anthropic$5.00$0.50$25.00$2.50 / $12.50200K$0.023$22.50$684.38$8,212.50
Claude Opus 4.6claude-opus-4-6Anthropic$5.00$0.50$25.00$2.50 / $12.501M$0.023$22.50$684.38$8,212.50
Claude Opus 4.7claude-opus-4-7Anthropic$5.00$0.50$25.00$2.50 / $12.501M$0.023$22.50$684.38$8,212.50
Claude Opus 4.8claude-opus-4-8Anthropic$5.00$0.50$25.00$2.50 / $12.501M$0.023$22.50$684.38$8,212.50
Claude Opus 5claude-opus-5Anthropic$5.00$0.50$25.00$2.50 / $12.501M$0.023$22.50$684.38$8,212.50
GPT-5.5gpt-5.5OpenAI$5.00$0.50$30.00$2.50 / $15.001.05M$0.025$25.00$760.42$9,125.00
Claude Fable 5claude-fable-5Anthropic$10.00$1.00$50.00$5.00 / $25.001M$0.045$45.00$1,368.75$16,425.00
Claude Fable 5.1claude-fable-5-1Anthropic$10.00$0.25$50.00$5.00 / $25.001M$0.045$45.00$1,368.75$16,425.00
GPT-6 Astragpt-6-astraOpenAI$10.00$1.00$50.00$5.00 / $25.001.05M$0.045$45.00$1,368.75$16,425.00
Daybreak Redgpt-daybreak-red-latestOpenAI$12.50$1.25$75.00-400K$0.063$62.50$1,901.04$22,812.50
o3-proo3-proOpenAI$20.00-$80.00$10.00 / $40.00200K$0.080$80.00$2,433.33$29,200.00
GPT-5 Progpt-5-proOpenAI$15.00-$120.00$7.50 / $60.00400K$0.090$90.00$2,737.50$32,850.00
GPT-5.2 Progpt-5.2-proOpenAI$21.00-$168.00$10.50 / $84.00400K$0.126$126.00$3,832.50$45,990.00
GPT-5.4 Progpt-5.4-proOpenAI$30.00-$180.00$15.00 / $90.001.05M$0.150$150.00$4,562.50$54,750.00
GPT-5.5 Progpt-5.5-proOpenAI$30.00-$180.00$15.00 / $90.001.05M$0.150$150.00$4,562.50$54,750.00

How the cost is worked out: per request = uncached input tokens × input price + cached input tokens × cached price + (output + reasoning tokens) × output price, all per 1M tokens. A day is that × requests per day, a month is 365 ÷ 12 days, a year is 365 days. Prompts over a model's long-context threshold use its long-prompt price. Cache write surcharges, cache storage, tool, search, image and audio fees are not included.

Pricing an AI feature? We design and build them.AI developmentTalk to us

Prices checked on 18 October 2026. Public list prices in US dollars per 1M tokens, copied by script from models.dev (MIT, list prices, cached-input prices, context windows) and LiteLLM model price file (MIT, batch api prices and reranker prices). Nothing is typed by hand. Confirm a price on the provider's own page before you commit to it; providers change prices and add surcharges (long prompts, regions, priority tiers) that a summary table cannot show.

Rules applied, and where we read them

Price changes

No price has moved since tracking began on 18 October 2026. Every refresh compares the new prices with the last snapshot and adds a dated line here for each one that changed.

How to read the table

List prices, your usage, and what moves the bill.

  • 01

    Input, cached and output prices

    Providers charge per token, and output usually costs several times more than input. A cached input token, the part of a prompt the provider has seen recently, is billed at a fraction of the full input price.

  • 02

    Batch and long prompts

    The Batch API trades speed for price: requests finish within hours instead of seconds. Some models charge more per token once a prompt passes a long-context threshold, and the calculator switches to that price for you.

  • 03

    Reasoning tokens count as output

    Reasoning models think before they answer, and those hidden tokens are billed as output. If your model reasons, add an estimate of reasoning tokens per request, or the monthly figure will be too low.

FAQs · LLM API pricing

LLM API pricing · frequently asked questions

Where the prices come from, how often they change and what the calculator leaves out.

A script copies them from two open, MIT-licensed price files: models.dev for list prices, cached-input prices and context windows, and LiteLLM for Batch API prices. A batch price is shown only where both files agree on the standard price. The date the prices were checked is printed beside the table.
The snapshot is refreshed by script, and our build refuses to pass once it is more than 14 days old. Every refresh compares the new prices with the previous ones and adds a dated entry to the price-change log on this page for each price that moved.
Cost per request is the uncached input tokens times the input price, plus the cached tokens times the cached price, plus output and reasoning tokens times the output price, all per million tokens. That is multiplied by requests per day, and a month is 365 divided by 12 days.
It is the share of each prompt the provider reads from its prompt cache instead of processing again, typically a long system prompt or document that repeats across requests. Those tokens are billed at the cached-input price. Models with no cached price in the data are charged the full input price and flagged.
Cache write surcharges and cache storage fees, web search and tool fees, image, audio and video tokens, regional or priority-tier surcharges, and any discount you have negotiated. Check the provider’s own pricing page before you commit to a budget.
Usually because real token counts differ from the averages entered. Count a real prompt with our token counter, check how many reasoning tokens your model actually uses in its response metadata, and re-run the calculator with those numbers.

Next step

Pricing an AI feature you plan to build?

QalbIT designs and builds AI features into web and mobile products: retrieval, agents and integrations with the model that fits your cost and quality target.

AI development

Keep exploring

More free tools

More free tools

Each one opens on a worked example and shows its answer without a signup.

All free tools

Further reading