ModelPriceWatch.com
Last scan 2026-08-09 Models tracked 198 Providers 30 Cheapest paid Granite 4.0 Micro $0.017/Mtok in Every price links to its source

Pricing trends

How LLM API pricing is evolving — price drops, provider competitiveness, and market movements over time. Built from twice-daily snapshots the scan never overwrites.

State of LLM pricing

market overview — computed from the current scan

Models tracked

178

current models — deprecated and preview tiers excluded

Providers

30

every price read from the provider's own pricing page

Cheapest paid · blended

$0.041 /Mtok

Granite 4.0 Micro — IBM

Most expensive · blended

$67.50 /Mtok

GPT-5.4 Pro — OpenAI

How prices are moving

movement — aggregated over a 90-day window, ranked by size

Repricing activity over the 90 days to Aug 1, 2026. For the full chronological log of every launch, price move and context change — plus JSON and Atom feeds — see the Newsroom.

Repricings in 90d

14

tracked list-price changes inside the window

Price drops

7

blended price went down — cheaper than before

Price increases

7

the moves that show up on invoices

Median move size

34.8%

half of the moves in the window were smaller than this

Biggest moves

Ranked by size of the move, not by date — the question the chronological feed can't answer.

The largest tracked LLM API price moves in the last 90 days — model, provider, direction, old and new blended, input and output prices per million tokens, size of the move, and date.
Model Move Blended $/1M In $/1M Out $/1M Size Date
Gemini 2.5 FlashGoogle increase $0.131$0.850 $0.075$0.300 $0.300$2.50 547.6% Jul 4, 2026
GPT-Image-2OpenAI increase $6.25$13.50 $5.00$8.00 $10.00$30.00 116% Jun 25, 2026
Gemini 2.5 FlashGoogle drop $0.850$0.131 $0.300$0.075 $2.50$0.300 84.6% Jun 25, 2026
GPT-5.6 LunaOpenAI drop $2.25$0.450 $1.00$0.200 $6.00$1.20 80% Aug 1, 2026
MiniMax-M3MiniMax drop $1.05$0.525 $0.600$0.300 $2.40$1.20 50% Jun 17, 2026
Gemini 3.1 Flash-LiteGoogle drop $1.01$0.563 $0.450$0.250 $2.70$1.50 44.4% Jun 9, 2026
Mistral Small 3.2 24BMistral increase $0.110$0.150 $0.080$0.100 $0.200$0.300 36.4% Jul 5, 2026
Pixtral 12BMistral drop $0.150$0.100 $0.150$0.100 $0.150$0.100 33.3% Jun 25, 2026
Drops are green, increases red — every move keeps its date and its old → new blended price, and snapshots are append-only, so history can't be rewritten. MODELPRICEWATCH.COM · 2026-08-09

Who reprices most

Providers by number of tracked price changes in the window — a proxy for how actively each one competes on price.

Mistral
5
OpenAI
4
Google
3
MiniMax
1
Together
1

Provider competitiveness

median blended $/Mtok per provider

Median blended cost per provider — lower means more competitive on price. The count is how many current models each provider lists.

Baichuan (1)
$0.070/M
Voyage AI (8)
$0.120/M
DeepInfra (4)
$0.150/M
Tencent (1)
$0.257/M
IBM (5)
$0.262/M
Upstage (1)
$0.262/M
Groq (6)
$0.365/M
Mistral (8)
$0.450/M
MiniMax (2)
$0.525/M
DeepSeek (2)
$0.544/M
Cohere (6)
$0.750/M
Z.AI (14)
$1.00/M
Alibaba (14)
$1.13/M
Fireworks (15)
$1.13/M
Meituan (1)
$1.30/M
Together (10)
$1.35/M
Amazon (4)
$1.40/M
Relace (2)
$1.50/M
xAI (4)
$1.56/M
Thinking Machines (2)
$1.76/M
Meta (5)
$2.00/M
Reka (3)
$3.00/M
Google (12)
$3.38/M
Moonshot (5)
$3.42/M
AI21 Labs (2)
$3.50/M
Perplexity (4)
$3.50/M
01.AI (1)
$4.50/M
OpenAI (24)
$4.81/M
Anthropic (11)
$10.00/M
Sakana AI (1)
$11.25/M

What we're tracking

coverage — how the history is built

Model Price Watch takes twice-daily snapshots of every model's pricing across all major LLM providers. Each snapshot records input price, output price, cached input price, context window, and status — so we capture not just what changed, but when and by how much.

Currently tracking 178 models across 30 providers. Snapshots are captured automatically twice a day by our pricing pipeline. Historical data is never overwritten — each snapshot is append-only, so every recorded price point is preserved.

Dig deeper into the token economy: