Pay as you go.

No subscription. Credits never expire.

Free to start: a one-time $1 credit for eligible Google, GitHub or Microsoft signups and 10 concurrent sessions before you pay anything. Top-ups from $5.

Browsers

$0.02 / hour

Residential proxy: $5 / GB

Browser agents

Model cost + 20%

Plus browser time and traffic.

At scale: 1,000 browser-hours cost $20, plus traffic. Compare cloud browsers

Get started Email us

Which model to choose?

Price vs. quality. Very hard browser tasks.

OpenAIAnthropicDeepSeekxAIMiniMaxZ.ai

Quality / 100 ↑

← Lower price per task (USD)

GPT-6 Luna · high · 53.9/100 · ≥ $0.15 / task

About this benchmark

Historical BU_Bench_v2 results cover 60 tasks per setting. Quality scores measure findings, not task success.

Builds differ. Recorded agent costs exclude judge, browser hosting, runner, and unrecorded calls. These are not Cloud prices.

Benchmark results for available models with comparable runs
Model	Quality / 100	USD / task	Minutes / task
Claude Opus 5	50.5	$7.371	35.9
Claude Opus 5.5	59.4	$4.215*	34.2
DeepSeek V4.1 Flash · high	41.5	$0.152*	16.3
DeepSeek V4.1 Flash · low	38.2	$0.138*	17.0
DeepSeek V4.1 Flash · max	47.9	$0.263*	31.8
DeepSeek V4.1 Flash · none	29.0	$0.118*	17.4
GLM 5.3 Flash	28.0	$0.114	28.8
GPT-5.6 Luna · default	16.1	$0.056	5.3
GPT-5.6 Luna · xhigh	41.2	$0.358	22.2
GPT-5.6 Sol · default	38.5	$2.715	13.2
GPT-5.6 Sol · xhigh	49.1	$7.803	23.6
GPT-6 Astra · medium	80.6	$8.867*	31.7
GPT-6 Luna · high	53.9	$0.155*	25.9
GPT-6 Luna · medium	50.3	$0.126*	19.8
GPT-6 Luna · none	31.8	$0.077*	8.7
GPT-6 Luna · xhigh	57.6	$0.190*	30.7
GPT-6 Sol · medium	66.8	$1.203*	13.5
Grok 4.5	33.0	$1.153	16.3
Grok 4.6	31.2	$3.934	23.4
MiniMax M3	11.8	$0.883	28.2

* Recorded cost is a lower bound. Model availability depends on your project.

No comparable result in this snapshot: BU Ultrafast, BU Fast, GPT-6.1 Sol, MiMo V2.6 Pro, MiMo V2.6 Flash.

GPT-6 Luna low has incomplete cost data and does not appear on the chart.

Benchmark ↗Source data ↗

Model prices

API V4 · USD per 1M tokens. Our 20% fee is already included.

Model prices in US dollars per one million tokens
Model	Input	Cached input	Output
Claude Opus 5	$6.00	$0.60	$30.00
Claude Sonnet 5	$2.40	$0.24	$12.00
Grok 4.5	$2.40	$0.36	$7.20
GPT-6 Astra	$12.00	$1.20	$60.00
GPT-5.6	$6.00	$0.60	$36.00
GPT-5.6 Luna	$0.24	$0.024	$1.44
Gemini 3.6 Flash	$1.80	$0.18	$9.00
MiniMax M3	$0.36	$0.072	$1.44

Claude Opus 5

Input: $6.00
Cached input: $0.60
Output: $30.00

Claude Sonnet 5

Input: $2.40
Cached input: $0.24
Output: $12.00

Grok 4.5

Input: $2.40
Cached input: $0.36
Output: $7.20

GPT-6 Astra

Input: $12.00
Cached input: $1.20
Output: $60.00

GPT-5.6

Input: $6.00
Cached input: $0.60
Output: $36.00

GPT-5.6 Luna

Input: $0.24
Cached input: $0.024
Output: $1.44

Gemini 3.6 Flash

Input: $1.80
Cached input: $0.18
Output: $9.00

MiniMax M3

Input: $0.36
Cached input: $0.072
Output: $1.44

Example: $1 in model tokens + 20% service fee → $1.20. Browser time and traffic are extra.

Your own key: pay your provider directly. Browser Use charges 20% of published token rates, including cached rates, plus browser time and traffic.

Provider-specific cache-write multipliers apply.

Concurrency tiers

Start with 10 concurrent sessions. Payments count in full as soon as the charge succeeds: one $1,000 top-up unlocks 250, $5,000 unlocks 500. Top-ups are credits that never expire.

Shared agent and browser concurrency by total paid
Total paid	Concurrent sessions
$0	10
$200	50
$1,000	250
$5,000	500
$25,000	1,000

Need more? Ask us.

Successful USD payments per project, net of refunds. Disputed payments and free credits don’t count. Your limit never goes down. Projects created before tiers launched, and AWS Marketplace subscriptions, keep their existing limits.

Billing details

Billing terms and usage charges in US dollars
Minimum top-up	$5 · unlocks every model
Credit expiry	Never
Free credits	$1 once per person. Eligible Google, GitHub or Microsoft signups. No card required.
Browser billing	Rounded up to whole minutes. 1-minute minimum.
High-quality residential proxies	$5 / GB · on by default. Routes traffic through another IP.
Direct / your own proxy	$0.20 / GB instead. Your proxy provider may charge separately.
Legacy subscriptions	No subscription plans are offered. Existing subscribers can manage billing ↗

Custom model API

Use our model with your own agent. Prices per 1M tokens.

Model prices in US dollars per one million tokens
Model	Input	Cached input	Output
BU 2.0	$0.60	$0.06	$3.50

BU 2.0

Input: $0.60
Cached input: $0.06
Output: $3.50

Legacy pricing