LLM benchmark for SEO: AI agent runs and model tests Data version 2026-11-22. Page: https://aiagentsforseo.net/stats Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/deed.en). Reuse with credit and a link to https://aiagentsforseo.net/stats. Cite as: Isac, D. (2026). LLM benchmark for SEO: AI agent runs and model tests (data version 2026-11-22) [Data set]. AI Agents for SEO. https://aiagentsforseo.net/stats. Files: https://aiagentsforseo.net/data/2026-11-22/agent-run-stats.json FILES agent-runs-by-role-and-model.csv runs on real work, one row per agent role and model (10 rows) agent-measured-jobs.csv single runs with exact cost, time and scope (8 rows) agent-model-tests.csv same-task model tests, one row per round, task and model (210 rows) agent-run-stats.json all of the above plus column definitions, methods and the judges' brief Frozen copies of this version: https://aiagentsforseo.net/data/2026-11-22/ RUN TABLES (agent-runs-by-role-and-model.csv, agent-measured-jobs.csv) Source: Takibi Base (https://takibibase.com), the agent workspace the author builds: the run records it keeps for agent runs on real client and own-site tasks. Rows: A filtered subset, not every run: 7 roles, only runs that recorded a model and an end time, and only role and model pairs with at least 3 such runs. The workspace's own role totals are larger (SEO Analyst: 152 runs in the workspace, 68 here). Two rows from the first setup days (runs that failed before the model did any work) are left out. Limit: The run tables are real tasks, not a controlled test. Each model got different tasks in a different period (roles changed default model over time), so the rows describe what each setup cost and how often it ended done, not which model is better at the same job. Done: Not done includes runs stopped by setup problems, limits or the person running them. Cost: Data $ is the connector call fees plus the paid data APIs billed through them (for example DataForSEO, Apollo). Model $ and data $ are separate medians per run. Their sum is a guide to a typical run's cost, not the median total cost of a run. MODEL TESTS (agent-model-tests.csv) The same-task model tests are controlled but small: 1 or 2 runs per model per task, scored by two LLM judges, so compare models within a round and read them as findings, not a ranking. - Few runs: 1 or 2 per model per task. One bad run moves a row a lot. - One or two tasks per role. A result says how a model did on that task, not on the role in general. - The judges are LLMs. They disagree at times (rows note where), and their exact model versions are not recorded. One judge, Muse, comes from the same vendor family as one contestant, Muse Ember. - Hosts: tencent/hy4-flash had a single host at the time (Tencent), whose content filter stopped one run; thinkingmachines/inkling-2-small had six hosts. Hosts for the other models are not recorded in the source. - Prices are those recorded on the dates shown (2026-11-07 to 2026-11-21) and will change. - The 2026-11-07 rounds have no judge scores in the source files; their rows carry cost, time, calls and end state only. - Tasks are modelled on real public software companies' websites, with made-up client notes. They are SEO and content-agency work for small SaaS products, not a general benchmark. - Podcast round (podcast-r1): data_usd includes page reads charged at the provider's credit ceiling. The real provider spend for that round's 5 runs was 910 credits, against a ceiling sum of 995, so the data figures for that round run high. Redaction: Client sites, people and task details are removed. Role names are generic job labels.