BlogLLM Benchmark Scores as Published, Side by Side

X Browse: Results as Published

Every result the catalog holds for X Browse: 4 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 xAI (internal)Introduced 2025Data version 2026103001, as of 2026-10-30

X Browse is an internal xAI benchmark that tests an agent's multi-hop search and browsing on X (the social network): the agent has to find and combine posts to answer a question. Its questions and scoring are not published.

X Browse

ModelValuePrinted bySetup as printed
GPT-5OpenAI, released 2025-08-0724.2xAI, Grok 4.1 Fast and Agent Tools APIchecked 2026-10-15Tools: agentic search (web search, X search, code execution); exact tools per row not stated
Claude Sonnet 4.5Anthropic, released 2025-09-2914.6xAI, Grok 4.1 Fast and Agent Tools APIchecked 2026-10-15Tools: agentic search (web search, X search, code execution); exact tools per row not stated
Gemini 3 ProGoogle, released 2025-11-1826.5xAI, Grok 4.1 Fast and Agent Tools APIchecked 2026-10-15Tools: agentic search (web search, X search, code execution); exact tools per row not stated
Grok 4.1 FastxAI, released 2025-11-1956.3xAI, Grok 4.1 Fast and Agent Tools APIchecked 2026-10-15Tools: agentic search (web search, X search, code execution); exact tools per row not stated

X Browse (average cost per task)

lower is better

ModelValuePrinted bySetup as printed
GPT-5OpenAI, released 2025-08-07$0.198xAI, Grok 4.1 Fast and Agent Tools APIchecked 2026-10-15Tools: agentic search (web search, X search, code execution); exact tools per row not stated
Claude Sonnet 4.5Anthropic, released 2025-09-29$0.126xAI, Grok 4.1 Fast and Agent Tools APIchecked 2026-10-15Tools: agentic search (web search, X search, code execution); exact tools per row not stated
Grok 4.1 FastxAI, released 2025-11-19$0.091xAI, Grok 4.1 Fast and Agent Tools APIchecked 2026-10-15Tools: agentic search (web search, X search, code execution); exact tools per row not stated

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.