data_analytics
Metric definition workshop
Turn ambiguous KPI definitions into precise, measurable specs.
#1 Recommendation
gpt-4o
Strong on DuckDB NSQL Leaderboard all_execution_accuracy (77%) and JSONSchemaBench Leaderboard medium_schema_compliance_pct (100%)
external/openai/gpt-4o
26.6%
Score
46.0%
Confidence
Limited benchmark evidence for this use case.
41 ranked models with average evidence of 7.8 points. Rankings may shift as more benchmark data is ingested.
Ranked Models
30
Evidence Quality
81%
Scoring
Benchmark-backed
Top Signal
DuckDB NSQL Leaderboard: all_execution_accuracy
All Ranked Models
| Rank | Model | Score |
|---|---|---|
| #1 | gpt-4o Strong on DuckDB NSQL Leaderboard all_execution_accuracy (77%) and JSONSchemaBench Leaderboard medium_schema_compliance_pct (100%) | 26.6% |
| #2 | qwen-2.5-72b-instruct Strong on DuckDB NSQL Leaderboard all_execution_accuracy (83%) and JSONSchemaBench Leaderboard medium_schema_compliance_pct (90%) | 25.1% |
| #3 | gpt-4o-20241120 Strong on DuckDB NSQL Leaderboard all_execution_accuracy (96%) and DuckDB NSQL Leaderboard hard_execution_accuracy (75%) | 23.8% |
| #5 | deepseek/deepseek-r1 | 21.5% |
| #10 | openai/gpt-4o-mini-2024-07-18 | 17.0% |
| #13 | gpt-4o-2024-08-06 | 16.0% |
| #20 | Llama-3.3-70B-Instruct | 13.8% |
| #25 | google/gemini-2.0-flash-001 | 13.2% |
| #28 | Qwen3-30B-A3B | 12.6% |
| #29 | Qwen2.5-Coder-7B | 12.5% |
| #30 | gemma-2-27b-it | 12.1% |
| #32 | phi-4 | 11.7% |
| #35 | Qwen3-32B | 11.1% |
| #39 | QwQ-32B-Preview | 10.8% |
| #41 | Qwen2.5-32B-Instruct | 10.5% |
| #42 | Phi-3-medium-128k-instruct | 10.5% |
| #45 | gemini-3-pro-preview | 9.2% |
| #46 | Meta-Llama-3.1-8B | 9.1% |
| #49 | gpt-4.1-20250414 | 8.9% |
| #50 | Llama-3.1-70B-Instruct | 8.8% |
| #52 | Phi-3-mini-128k-instruct | 8.7% |
| #53 | gemini-2.5-pro | 8.7% |
| #54 | Grok-4-0709 | 8.6% |
| #56 | claude-sonnet-4-20250514 | 8.3% |
| #58 | Qwen2.5-14B-Instruct | 8.1% |
| #62 | gemini-2.5-flash | 7.3% |
| #64 | Meta-Llama-3-8B-Instruct | 7.2% |
| #65 | Llama-3.1-8B-Instruct | 7.1% |
| #67 | deepseek-v3 | 6.8% |
| #71 | Qwen2.5-Coder-3B-Instruct | 6.6% |
Compare Models
Model A leads by +1.5%
Shareable Link →Model A
gpt-4o
external/openai/gpt-4o
Rank #1
DuckDB NSQL Leaderboard: all_execution_accuracy
Value 76.9% · Conf 100.0% · Weight 8.1%
duckdb_nsql_leaderboard.all_execution_accuracy (Mar 12, 2026)
JSONSchemaBench Leaderboard: medium_schema_compliance_pct
Value 100.0% · Conf 100.0% · Weight 5.0%
jsonschemabench_leaderboard.medium_schema_compliance_pct (Mar 12, 2026)
JSONSchemaBench Leaderboard: hard_schema_compliance_pct
Value 100.0% · Conf 100.0% · Weight 3.4%
jsonschemabench_leaderboard.hard_schema_compliance_pct (Mar 12, 2026)
DuckDB NSQL Leaderboard: hard_execution_accuracy
Value 50.0% · Conf 100.0% · Weight 2.2%
duckdb_nsql_leaderboard.hard_execution_accuracy (Mar 12, 2026)
Model B
qwen-2.5-72b-instruct
external/qwen/qwen-2-5-72b-instruct
Rank #2
DuckDB NSQL Leaderboard: all_execution_accuracy
Value 82.7% · Conf 100.0% · Weight 8.8%
duckdb_nsql_leaderboard.all_execution_accuracy (Mar 12, 2026)
JSONSchemaBench Leaderboard: medium_schema_compliance_pct
Value 90.1% · Conf 100.0% · Weight 4.5%
jsonschemabench_leaderboard.medium_schema_compliance_pct (Mar 12, 2026)
JSONSchemaBench Leaderboard: hard_schema_compliance_pct
Value 74.4% · Conf 100.0% · Weight 2.5%
jsonschemabench_leaderboard.hard_schema_compliance_pct (Mar 12, 2026)
Galileo Agent Leaderboard v2: Avg AC
Value 76.1% · Conf 100.0% · Weight 1.2%
galileo_agent_v2.avg_ac (Mar 12, 2026)
▶Ranking Diagnostics & Missing Models
Source Lift
Ranked
41
Sources
8
Quality
Insufficient
DuckDB NSQL Leaderboard
duckdb_nsql_leaderboard
23 rows
4.0% avg lift
JSONSchemaBench Leaderboard
jsonschemabench_leaderboard
12 rows
2.4% avg lift
EQ-Bench Leaderboard
eq_bench
10 rows
0.4% avg lift
Vals Legal Bench
vals_legal_bench
10 rows
0.4% avg lift
Missing Strong Models
anthropic/claude-sonnet-4.6
external/anthropic/claude-sonnet-4-6
Rank #4
21.1%
gpt-5-mini-2025-08-07
external/openai/gpt-5-mini-2025-08-07
Rank #7
19.6%
google/gemini-3.1-pro-preview
external/google/gemini-3-1-pro-preview
Rank #8
19.3%
gpt-5-2025-08-07
external/openai/gpt-5-2025-08-07
Rank #9
19.2%
▶Taxonomy Details
Core Tasks
Required Modes
Domains
Related Use Cases
data_analytics
SQL debugging
Diagnose and fix SQL queries for correctness and performance.
Top: gpt-4o-20241120
data_analytics
Dashboard narratives
Generate weekly KPI narratives and investigation suggestions.
Top: gemini-3-pro-preview
data_analytics
Text-to-SQL analyst assistant
Convert questions into SQL and explain the query.
Top: gemini-3-pro-preview
data_analytics
Insight mining from text corpora
Extract themes and actions from large text datasets.
Top: qwen-2.5-72b-instruct