developer_tools
Refactoring assistant
Refactor code safely while preserving behavior and improving clarity.
#1 Recommendation
z-ai/glm-4.7
Strong on Sonar Java Quality Leaderboard functional_skill_pct (74%) and Vals LiveCodeBench overall_accuracy_pct (91%)
external/z-ai/glm-4-7
21.8%
Score
31.4%
Confidence
Limited benchmark evidence for this use case.
53 ranked models with average evidence of 12.8 points. Rankings may shift as more benchmark data is ingested.
Ranked Models
30
Evidence Quality
81%
Scoring
Benchmark-backed
Top Signal
Sonar Java Quality Leaderboard: functional_skill_pct
All Ranked Models
Compare Models
Model A leads by +0.5%
Shareable Link →Model A
z-ai/glm-4.7
external/z-ai/glm-4-7
Rank #1
Sonar Java Quality Leaderboard: functional_skill_pct
Value 74.4% · Conf 100.0% · Weight 1.9%
sonar_java_quality.functional_skill_pct (Mar 12, 2026)
Vals LiveCodeBench: overall_accuracy_pct
Value 91.4% · Conf 100.0% · Weight 1.4%
vals_lcb.overall_accuracy_pct (Mar 12, 2026)
Vals SWE-bench: overall_accuracy_pct
Value 79.9% · Conf 100.0% · Weight 1.4%
vals_swebench.overall_accuracy_pct (Mar 12, 2026)
Sonar Java Quality Leaderboard: issue_density_error_per_kloc
Value 65.2% · Conf 100.0% · Weight 1.3%
sonar_java_quality.issue_density_error_per_kloc (Mar 12, 2026)
Model B
Kimi K2 Thinking
external/kimi/kimi-k2-thinking
Rank #2
Sonar Java Quality Leaderboard: functional_skill_pct
Value 88.4% · Conf 100.0% · Weight 2.2%
sonar_java_quality.functional_skill_pct (Mar 12, 2026)
Sonar Java Quality Leaderboard: issue_density_error_per_kloc
Value 66.6% · Conf 100.0% · Weight 1.3%
sonar_java_quality.issue_density_error_per_kloc (Mar 12, 2026)
Vals SWE-bench: overall_accuracy_pct
Value 63.5% · Conf 100.0% · Weight 1.1%
vals_swebench.overall_accuracy_pct (Mar 12, 2026)
Vals LiveCodeBench: overall_accuracy_pct
Value 65.1% · Conf 100.0% · Weight 1.0%
vals_lcb.overall_accuracy_pct (Mar 12, 2026)
▶Ranking Diagnostics & Missing Models
Source Lift
Ranked
53
Sources
8
Quality
Insufficient
Vals LiveCodeBench
vals_lcb
39 rows
1.1% avg lift
Vals Legal Bench
vals_legal_bench
39 rows
0.3% avg lift
Vals Tax Eval v2
vals_tax_eval_v2
38 rows
0.3% avg lift
Vals GPQA
vals_gpqa
38 rows
0.2% avg lift
Missing Strong Models
google/gemini-2.0-flash-001
external/google/gemini-2-0-flash-001
Rank #56
10.3%
deepseek-v3
external/deepseek-ai/deepseek-v3
Rank #66
8.8%
Llama-4-Scout-17B-16E-Instruct
external/meta/llama-4-scout-17b-16e-instruct
Rank #73
8.4%
▶Taxonomy Details
Core Tasks
Required Modes
Domains
Related Use Cases
developer_tools
Codebase onboarding brief
Summarize a repository's architecture, modules, and conventions.
Top: gemini-3-pro-preview
developer_tools
Code generation
Generate correct, secure code from requirements.
Top: anthropic/claude-sonnet-4.6
developer_tools
PR review agent
Review diffs for correctness, security, and maintainability.
Top: anthropic/claude-sonnet-4.6
developer_tools
Agentic bug fixing
Agentic loop that reproduces, fixes, and validates with tests.
Top: anthropic/claude-sonnet-4.6