LLM Pricing Tracker: API and Subscription Costs

Mar 19, 2026

A tracker for leading LLM API token prices and consumer subscriptions, with official links, repo snapshot refreshes, price-history charts, and live benchmark/provider snapshots.

Note

Latest repo snapshot in this build: May 7, 2026. Click Check for newer snapshot below to query GitHub for a fresher snapshot. Each browser is limited to one remote check per day.

This page tracks public pricing from official provider pages for major frontier-model vendors I regularly compare: OpenAI, Google (Gemini and Gemma), Anthropic, xAI, DeepSeek, Qwen, Moonshot/Kimi, Xiaomi/MiMo, MiniMax, Together AI (including GLM-5 and public Llama endpoints), Meta/Llama references, and GitHub Copilot.

A few quick cautions before using the numbers:

  • API pricing and consumer subscription pricing are different products.
  • Some vendors publish tiered pricing by context length, region, or prompt type.
  • When a vendor does not publicly expose a comparable token-billing number, I mark that clearly instead of guessing.
  • The charts below use snapshots stored in this repo, including daily-refreshable Artificial Analysis benchmark/provider snapshots and manually curated pricing rows from official vendor pages.

Latest loaded snapshot: May 7, 2026
This tracker loads the latest repo snapshot. The button below checks GitHub for a newer snapshot at most once per day in this browser; official pricing pages can still change between repo refreshes.

API Token Fees

Representative API pricing for major frontier-model vendors. Numbers are taken from official vendor pricing pages; mixed currencies are kept in the vendor’s quoted currency.

VendorModel / ProductInputCached inputOutputNotesOfficial
OpenAIGPT-5.5
USD per 1M tokens
$5.00$0.50$30.00OpenAI's pricing page now lists GPT-5.5 as the latest flagship model. The page labels it coming soon and notes higher long-context multipliers above 270K input tokens.OpenAI API pricing
GoogleGemini 2.5 Pro
USD per 1M tokens
$1.25 (<=200K prompts)$0.125 (<=200K prompts)$10.00 (<=200K prompts)Tiered pricing. For prompts above 200K tokens, Google lists $2.50 input, $0.25 cached input, and $15.00 output.Gemini API pricing
GoogleGemma 4
USD per 1M tokens
Free of chargeFree of chargeFree of chargeGoogle's current Gemini Developer API pricing page lists Gemma 4 with free input, output, and context caching, while the paid tier remains unavailable.Google Gemini Developer API pricing
AnthropicClaude Sonnet 4 / Claude Code backend
USD per 1M tokens
$3.00$0.30 (cache hits)$15.00Claude Code team/API usage is billed from Claude API token consumption. Anthropic also lists separate cache-write prices.Anthropic model pricing
DeepSeekdeepseek-v4-flash
USD per 1M tokens
$0.14 (<=128K cache miss)$0.028 (<=128K cache hit)$0.28 (<=128K)Official DeepSeek V4 Flash overseas pricing. For prompts above 128K tokens, DeepSeek lists $0.28 input, $0.056 cached input, and $0.56 output per 1M tokens.DeepSeek models & pricing
DeepSeekdeepseek-v4-pro
USD per 1M tokens
$1.74 (<=128K cache miss)$0.145 (<=128K cache hit)$3.48 (<=128K)Official DeepSeek V4 Pro overseas pricing. For prompts above 128K tokens, DeepSeek lists $3.48 input, $0.29 cached input, and $6.96 output per 1M tokens.DeepSeek models & pricing
Qwen / Alibaba Cloudqwen-max-latest
USD per 1M tokens
$1.60-$6.40Alibaba Cloud lists qwen-max-latest with non-thinking pricing and no tiered pricing on the current pricing page.Alibaba Cloud Model Studio pricing
Together AI / Z AIGLM-5
USD per 1M tokens
$1.00-$3.20Together AI's public serverless price for Z AI's GLM-5. The public model page does not list a separate cached-input rate.Together AI GLM-5 pricing
Meta / Llama via Together AILlama 4 Maverick
USD per 1M tokens
$0.27-$0.85Meta's Llama 4 Maverick served through Together AI's public serverless API. Meta's public developer docs describe the model family, while Together AI exposes a comparable public token price.Together AI Llama 4 Maverick pricing
Moonshot AI / Kimikimi-latest
CNY per 1M tokens
¥2 / ¥5 / ¥10 (8k/32k/128k)¥1.00 (auto cache hit)¥10 / ¥20 / ¥30 (8k/32k/128k)Moonshot's official April 7, 2025 pricing notice shows kimi-latest auto-selects the 8K / 32K / 128K tier at ¥2 / ¥5 / ¥10 input and ¥10 / ¥20 / ¥30 output per 1M tokens. Automatic cache-hit billing remains ¥1 / 1M tokens.Moonshot official pricing update
Xiaomi / MiMoMiMo-V2.5-Pro
USD per 1M tokens
$1.00 (<=256K cache miss)$0.20 (<=256K cache hit)$3.00 (<=256K)Official overseas pricing for Xiaomi's flagship MiMo-V2.5-Pro / MiMo-V2-Pro tier. For prompts above 256K tokens, Xiaomi lists $2.00 input, $0.40 cached input, and $6.00 output per 1M tokens.Xiaomi MiMo pricing
MiniMaxMiniMax-M2.5
CNY per 1M tokens
¥2.10¥0.21 (cache read)¥8.40Current MiniMax pay-as-you-go text pricing. Cache writes are listed separately at ¥2.625 / 1M tokens.MiniMax pay-as-you-go pricing
xAIgrok-4.20-beta-0309-reasoning
USD per 1M tokens
$2.00-$6.00xAI lists Grok 4.20 reasoning and non-reasoning variants with the same token pricing on the public API page.xAI API pricing
GitHub CopilotCopilot product pricing
N/A
N/AN/AN/AGitHub Copilot is sold as a subscription product. GitHub does not publicly publish a Copilot per-token API price comparable to the other vendors here.GitHub Copilot plans

Subscription Plans

Publicly listed consumer or team subscriptions from the official provider pages I checked for this tracker.

VendorPlanPriceNotesOfficial
OpenAIChatGPT Plus$20 / monthConsumer plan. API usage is billed separately.ChatGPT pricing
OpenAIChatGPT Pro$200 / monthHighest individual-access plan.ChatGPT pricing
OpenAIChatGPT Business$25 / user / month billed annually or $30 monthlyShared workspace for teams and growing businesses.ChatGPT pricing
AnthropicClaude Pro$20 / monthIndividual paid plan for Claude.Anthropic pricing
AnthropicClaude Max 5x$100 / month5x more usage than Pro. Includes Claude Code.Claude Max
AnthropicClaude Max 20x$200 / month20x more usage than Pro. Includes Claude Code.Claude Max
AnthropicClaude Team$25 / seat / month billed annually, $30 billed monthlyStandard team seats. Anthropic also lists premium seats at $150 / member / month, including Claude Code and higher usage.Claude Team billing
GoogleGoogle AI Pro$19.99 / monthFormerly AI Premium. Consumer plan with Gemini app, Flow, NotebookLM and more.Google AI plans
GoogleGoogle AI Ultra$249.99 / monthHighest Google AI subscription tier for the Gemini app and related tools.Google AI plans
Xiaomi / MiMoToken Plan (monthly)$6 / $16 / $50 / $100 per monthOfficial monthly Token Plan tiers: Lite, Standard, Pro, and Max. The package covers MiMo-V2.5-Pro, MiMo-V2.5, and the rest of Xiaomi's current V2 / V2.5 programming-tool lineup.MiMo Token Plan subscription
Xiaomi / MiMoToken Plan (annual)$63.36 / $168.96 / $528.00 / $1056.00 per yearOfficial annual Token Plan tiers: Lite, Standard, Pro, and Max.MiMo Token Plan subscription
MiniMaxCoding Plan Starter¥29 / month40 prompts every 5 hours.MiniMax Coding Plan
MiniMaxCoding Plan Plus¥49 / month100 prompts every 5 hours.MiniMax Coding Plan
MiniMaxCoding Plan Max¥119 / month300 prompts every 5 hours.MiniMax Coding Plan
xAI / GrokX Premium+$40 / month in the US web pricing tableHelp.x.com ties Premium+ to expanded Grok access. Regional prices vary.X Premium pricing
GitHub CopilotCopilot Pro$10 / month or $100 / yearIndividual developer plan.GitHub Copilot plans
GitHub CopilotCopilot Pro+$39 / month or $390 / yearHigher premium-request limits and broader model access.GitHub Copilot plans
GitHub CopilotCopilot Business$19 / seat / monthOrganization-managed plan from GitHub Docs.GitHub Docs
GitHub CopilotCopilot Enterprise$39 / seat / monthEnterprise-managed plan from GitHub Docs.GitHub Docs

Some vendors in the API table are omitted here because I could not find a public official subscription plan for them in the sources checked for this snapshot: DeepSeek, Qwen / Alibaba Cloud, Moonshot AI / Kimi.

Price History

Switch provider, metric, and time grain to compare the official pricing checkpoints I have stored so far.

Loading chart…

History lines use official pricing snapshots that are stored in this repo. Some providers only have one official snapshot recorded so far, while others mix successive flagship or promo models.

Artificial Analysis Benchmark Snapshot

The current top 10 models on the Artificial Analysis Intelligence Index. This list is intentionally model-level, so repeated vendors can appear more than once when they occupy multiple top-10 slots.

Loading benchmark snapshot…

Top 10 current models by Artificial Analysis Intelligence Index. Prices and speeds below come from Artificial Analysis' benchmark snapshot, so they may differ from the manual API rows above when multiple deployments or reasoning modes exist. Source: Artificial Analysis models leaderboard (checked May 7, 2026).

VendorBenchmark modelIntelligenceSpeedBlended pricePrompt pricingDetails
OpenAIGPT-5.5 (xhigh)
GPT-5.5 (xhigh)
60.2464.83 t/s$Input $5 · Output $30Model details
OpenAIGPT-5.5 (high)
GPT-5.5 (high)
58.8766.23 t/s$Input $5 · Output $30Model details
AnthropicClaude Opus 4.7 (max)
Claude Opus 4.7 (Adaptive Reasoning, Max Effort)
57.2845.24 t/s$Input $6.25 · Output $25Model details
GoogleGemini 3.1 Pro Preview
Gemini 3.1 Pro Preview
57.18125.55 t/s$Input $2 · Output $12Model details
OpenAIGPT-5.5 (medium)
GPT-5.5 (medium)
56.7160.44 t/s$Input $5 · Output $30Model details
KimiKimi K2.6
Kimi K2.6
53.930.06 t/s$Input $0.95 · Output $4Model details
XiaomiMiMo-V2.5-Pro
MiMo-V2.5-Pro
53.8357.27 t/s$Input $1 · Output $3Model details
OpenAIGPT-5.3 Codex (xhigh)
GPT-5.3 Codex (xhigh)
53.5678.73 t/s$Input $1.75 · Output $14Model details
xAIGrok 4.3
Grok 4.3
53.280.14 t/s$Input $1.25 · Output $2.5Model details
MetaMuse Spark
Muse Spark
52.150 t/s$Input $0 · Output $0Model details

Top-25 API Provider Leaderboard

Artificial Analysis provider leaderboard snapshot, aggregated to the best currently benchmarked endpoint for each provider. Switch metrics to compare intelligence, blended price, latency, and context window separately.

Loading provider leaderboard…

Each provider rank uses that provider's best currently benchmarked endpoint for the selected metric, so the representative model can change across metrics. Source: Artificial Analysis provider leaderboard (checked May 7, 2026).

Scale and Price Frontier

A Star History-style yearly line for open-weight model sizes and the highest reviewed output-token prices since 2021.

Loading scale and price frontier…

Lines start in 2021 and combine curated source-backed historical records with the latest Artificial Analysis model metadata. Model size tracks the largest open-weight/open-access LLM by total disclosed parameters for each year, so sparse MoE and dense models are not quality-equivalent. Output-token price uses public USD text-token API prices and excludes tool-call, image, audio, video, and subscription pricing. Source: Artificial Analysis models (checked May 7, 2026).