BlogLLM Benchmark Scores as Published, Side by Side

LiveSQLBench: Results as Published

Every result the catalog holds for LiveSQLBench: 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 Code GenerationMaintainer BIRD Team, HKU, and Google CloudIntroduced 2025Data version 2026103001, as of 2026-10-30

LiveSQLBench is a contamination-free text-to-SQL benchmark that adds new databases and tasks over time. The Base-Full v1 release, dated 4 September 2025, has 600 tasks on 22 real databases, covering business-intelligence queries and database management statements, with a hierarchical knowledge base for each database. A SELECT answer is checked by comparing its execution result with the gold SQL, and a management statement is checked with test cases. The score is the success rate, the share of tasks that pass those checks.

LiveSQLBench (Base-Full v1)

ModelValuePrinted bySetup as printed
GPT-5.5OpenAI, released 2026-04-2340.17%MiniMax, MiniMax M3checked 2026-10-15Harness: Claude Code
Claude Opus 4.7Anthropic, released 2026-04-1641%MiniMax, MiniMax M3checked 2026-10-15Harness: Claude Code
Gemini 3.1 Pro PreviewGoogle, released 2026-02-1939.83%MiniMax, MiniMax M3checked 2026-10-15Harness: Claude Code
MiniMax M3MiniMax, released 2026-06-0140.17%MiniMax, MiniMax M3checked 2026-10-15Harness: Claude Code
MiniMax M2.7MiniMax, released 2026-03-1833.17%MiniMax, MiniMax M3checked 2026-10-15Harness: Claude Code

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.