03Inference· as of 2026-07-06

What a million tokens costs

List prices from official provider pages, refreshed 2026-07-06. Commodity intelligence keeps deflating; the frontier premium is back.

Cheapest GPT-4-class · blended

$0.18/1M

−99.5% since Mar ’23 · DeepSeek V4 Flash

OpenAI flagship · GPT-5.5

$11.25/1M

Premium tier reset in Apr 2026

Cheapest frontier-tier · Grok 4.3

$1.56/1M

$1.25 in / $2.50 out, 1M context

The price of GPT-4-class intelligence

US$ per 1M tokens, blended 3:1 · log scale

Data — endpoints, license, and freshness behind The price of GPT-4-class intelligence

/llms-full.txt

Full plain-text dataset for answer engines: current GPU quotes, the forward board, token prices, accelerator specs, and citable class aggregates inlined as tables, each section stamped by its owning clock (market snapshot, live quotes, or core data).

curl -s https://aicomputetracker.com/llms-full.txt

No API key, CORS open (Access-Control-Allow-Origin: *) — fetch it from a browser, a notebook, or a cron job.

Licensed CC BY 4.0 — attribution “AI Compute Tracker (aicomputetracker.com), CC BY 4.0”.

Core data

Full data contract →

Cite — copy a citation for The price of GPT-4-class intelligence

Copy-ready citation

Cite as: AI Compute Tracker, “The price of GPT-4-class intelligence”, core data as of 2026-07-23, https://aicomputetracker.com/inference/#token-price-collapse, AI Compute Tracker (aicomputetracker.com), CC BY 4.0, accessed <date you visit>

Replace “<date you visit>” with your access date.

Inference-price chart awaiting interactive loadStatic chart frame · exact values in table view

Official pricing pages & launch announcements. ‘GPT-4-class’ = matches original GPT-4 on MMLU (~86.4).

Current list prices, blended

Every tracked model, July 2026 — dot position on a log scale.

OpenAIAnthropicGooglexAIDeepSeek
  • GPT-5.5 ProDetails for GPT-5.5 Pro

    Research-grade tier

    OpenAI
    $67.50
  • Claude Fable 5Anthropic
    $20.00
  • GPT-5.5OpenAI
    $11.25
  • Claude Opus 4.8Anthropic
    $10.00
  • GPT-5.4OpenAI
    $5.63
  • Gemini 3.1 ProDetails for Gemini 3.1 Pro

    ≤200K context tier

    Google
    $4.50
  • Claude Sonnet 5Details for Claude Sonnet 5

    Intro pricing; $3/$15 from Sep 2026

    Anthropic
    $4.00
  • Gemini 3.5 FlashGoogle
    $3.38
  • Claude Haiku 4.5Anthropic
    $2.00
  • GPT-5.4 miniOpenAI
    $1.69
  • Grok 4.3Details for Grok 4.3

    1M context, native video input

    xAI
    $1.56
  • Gemini 3.1 Flash-LiteGoogle
    $0.56
  • DeepSeek V4 ProDetails for DeepSeek V4 Pro

    Open weights (MIT)

    DeepSeek
    $0.55
  • GPT-5.4 nanoOpenAI
    $0.46
  • DeepSeek V4 FlashDetails for DeepSeek V4 Flash

    Cache-hit input $0.0028

    DeepSeek
    $0.18

Blended price = (3 × input + output) ÷ 4, per 1M tokens, standard list. Dot position uses the logarithmic scale above.

All tracked models

Standard (non-batch) list prices from official pricing pages, 2026-07-06.

Current LLM API prices per million tokens
GPT-5.5 Pro
Pricing note for GPT-5.5 Pro

Research-grade tier

OpenAI$30.00$180.00$67.50
Claude Fable 5
Anthropic$10.00$50.00$20.00
GPT-5.5
OpenAI$5.00$30.00$11.25
Claude Opus 4.8
Anthropic$5.00$25.00$10.00
GPT-5.4
OpenAI$2.50$15.00$5.63
Gemini 3.1 Pro
Pricing note for Gemini 3.1 Pro

≤200K context tier

Google$2.00$12.00$4.50
Claude Sonnet 5
Pricing note for Claude Sonnet 5

Intro pricing; $3/$15 from Sep 2026

Anthropic$2.00$10.00$4.00
Gemini 3.5 Flash
Google$1.50$9.00$3.38
Claude Haiku 4.5
Anthropic$1.00$5.00$2.00
GPT-5.4 mini
OpenAI$0.75$4.50$1.69
Grok 4.3
Pricing note for Grok 4.3

1M context, native video input

xAI$1.25$2.50$1.56
Gemini 3.1 Flash-Lite
Google$0.25$1.50$0.56
DeepSeek V4 Pro
Pricing note for DeepSeek V4 Pro

Open weights (MIT)

DeepSeek$0.44$0.87$0.55
GPT-5.4 nano
OpenAI$0.20$1.25$0.46
DeepSeek V4 Flash
Pricing note for DeepSeek V4 Flash

Cache-hit input $0.0028

DeepSeek$0.14$0.28$0.18

What does GPT-4-class inference cost today?

As of 2026-07-06, the cheapest model matching original-GPT-4 quality is DeepSeek V4 Flash at $0.18 per million tokens blended 3:1 — −99.5% versus GPT-4 at launch ($37.50 blended, Mar 2023). These are list prices: batch endpoints run about 50% off and cached input about 90% off.