One agent harness. A choice of models. Find the right balance for the work you actually run.
Managed agent models
Per 1M tokens · USD
| Model | Input | Output |
|---|---|---|
| BU 4.0 MiniRecommended | $0.20 | $1.00 |
| BU 4.0New US default | $1.80 | $9.00 |
| BU 4.0 ProRecommended | $3.00 | $15.00 |
| BU UltrafastRecommended | $0.24 | $1.20 |
| BU FastRecommended | $0.12 | $0.60 |
| GPT-6.1 SolRecommended | $2.40 | $12.00 |
| GPT-6 LunaEU + older default | $0.12 | $0.60 |
| GPT-6.1 Terra | $1.20 | $6.00 |
| Claude Opus 5.5 | $4.80 | $24.00 |
| Claude Sonnet 5.5 | $2.40 | $12.00 |
| GPT-6 Astra | $12.00 | $60.00 |
| GLM 5.3 Flash | $0.18 | $0.60 |
| Grok 4.6 | $2.40 | $7.20 |
| MiniMax M3 | $0.36 | $1.44 |
Omit the model and US projects created since November 16, 2026 run BU 4.0. Change the model as your task's accuracy, latency, and cost requirements change.
Model documentationEligible plans support your model-provider key. Provider token costs, Browser Use orchestration, browser time, and traffic remain separate charges.
Understand the chargesBrowser Use also develops models trained for browser work. BU 2.0 has its own model API rates, separate from managed V4 runs.
Purpose-built model pricingChoose with evidence
Accuracy and cost depend on the task, model, and harness. Start with a matched benchmark, then test your own workflow.
Model leaderboard