New: Agent Index for knowledge work is live.Explore the index ↗

Best LLMs for browser workflows

LiveBench: Browser Use

LiveBench 2.0 evaluates browser workflows on its October 25, 2026 release. This suite contains 25 tasks or questions, scored from 0 to 100 in one run per model. Models use their highest supported effort unless a different setting is reported. This is an individual task-suite score, separate from LiveBench’s weighted overall average.

Updated: Data as of:
Widely usedNot saturated
Official benchmark
LiveBench: Browser Use results. Each row identifies its benchmark version and evaluation track. Scores from different versions are not directly comparable. Best effort from the latest stored report date per model and version.
Model / creatorVersionReported score
GPT-6.1 SolOpenAI · Proprietary2.0 / 2026-10-2593.3%
GPT-6.1 TerraOpenAI · Proprietary2.0 / 2026-10-2593.1%
GPT-6 AstraOpenAI · Proprietary2.0 / 2026-10-2591.3%
Claude Fable 5.1Anthropic · Proprietary2.0 / 2026-10-2589.8%
Claude Sonnet 5.5Anthropic · Proprietary2.0 / 2026-10-2589.3%
Claude Opus 5.5Anthropic · Proprietary2.0 / 2026-10-2588.4%
Gemini 4 ArgonGoogle · Proprietary2.0 / 2026-10-2582.8%
MiniMax M3.1 FlashMiniMax · Open source / open weights2.0 / 2026-10-2580.6%
MiMo-V2.6-ProXiaomi · Open source / open weights2.0 / 2026-10-2577.3%
Kimi K3.1 ProMoonshot AI · Open source / open weights2.0 / 2026-10-2575.9%
Kimi K3Moonshot AI · Open source / open weights2.0 / 2026-10-2575.2%
GPT-5.6 TerraOpenAI · Proprietary2.0 / 2026-10-2570.3%
GPT-6 LunaOpenAI · Proprietary2.0 / 2026-10-2570.3%
Claude Haiku 5.5Anthropic · Proprietary2.0 / 2026-10-2569.5%
DeepSeek V4.1 Flash 1017DeepSeek · Open source / open weights2.0 / 2026-10-2567.2%
Muse Wonder 1.0Meta · Proprietary2.0 / 2026-10-2560%
GLM-5.3-FlashZ.ai · Open source / open weights2.0 / 2026-10-2559.8%
Grok 4.7xAI · Proprietary2.0 / 2026-10-2549.3%
Gemini 3.8 FlashGoogle · Proprietary2.0 / 2026-10-2545.1%
Muse Spark 1.3Meta · Proprietary2.0 / 2026-10-2530.9%
Muse Spark OpenMeta · Open source / open weights2.0 / 2026-10-2526.2%
Step 5 PreviewStepFun · Open source / open weights2.0 / 2026-10-2524.8%
Qwen3.8 27BQwen · Open source / open weights2.0 / 2026-10-2524.2%
Nemotron 3.5 Super VLNVIDIA · Open source / open weights2.0 / 2026-10-2521.6%
InklingThinking Machines · Open source / open weights2.0 / 2026-10-2512%
GLM-5.3Z.ai · Open source / open weights2.0 / 2026-10-140%
View all 26 published reports
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
Kimi K375.2%
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-25

release: 2026-10-25 · Highest supported effort unless publisher reports an exception; single run · suite: browser-use

LiveBench ↗
2.0 / 2026-10-14

release: 2026-10-14 · Official LiveBench 2.0 row GLM-5.3. One run per suite; effort not explicitly labeled on this row. Historical release, not the October 25 index track. · source model id: glm-5.3 · suite: browser-use

LiveBench ↗