BlogLLM Benchmark Scores as Published, Side by Side
WebVoyager: Results as Published
Every result the catalog holds for WebVoyager: 5 models from 4 developers, read from 1 page. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area Web Search and ResearchMaintainer Hongliang He and co-authors (Zhejiang University and Tencent AI Lab)Introduced 2024Data version 2026103001, as of 2026-10-30
WebVoyager is a web-agent benchmark of 643 tasks on 15 popular real websites (shopping, travel, maps, search and others): the agent browses from screenshots and page text, and a judge model checks whether the task was completed. The score is the task success rate.
Not checked against a maintainer's page.
WebVoyager
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| Qwen3.5-122B-A10BAlibaba, released 2026-02-24 | 54.89% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Qwen3.8-Flash-NextAlibaba | 90.39% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| GLM-5.3-FlashZ.ai, released 2026-08-26 | 72.64% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Nemotron 3.5 Super VL 120B-A12BNVIDIA, released 2026-10-15 | 77.72% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
| Inkling-SmallThinking Machines | 5.25% | NVIDIA, NVIDIA Nemotron 3.5 Super VL 120B-A12B model cardchecked 2026-10-19 | - |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.