BasedAGIBasedAGI
Menu
Rankings live

developer_tools

Documentation from code

Generate docstrings and technical docs that match code behavior.

#1 Recommendation

gpt-4.1-20250414

Strong on Galileo Agent Leaderboard v2 Avg AC (100%) and OpenVLM OCRBench Official ocrbench_score_pct (88%)

external/openai/gpt-4-1-20250414

19.1%

Score

30.1%

Confidence

Runners-up:#2 gemini-3-pro-preview (16.8%)#3 gemini-2.5-pro (16.2%)#4 deepseek/deepseek-r1 (15.7%)

Limited benchmark evidence for this use case.

48 ranked models with average evidence of 14.0 points. Rankings may shift as more benchmark data is ingested.

Ranked Models

30

Evidence Quality

79%

Scoring

Benchmark-backed

Top Signal

Galileo Agent Leaderboard v2: Avg AC

All Ranked Models

Max params:
Min confidence:
30 of 30
RankModelScore
#1gpt-4.1-20250414

Strong on Galileo Agent Leaderboard v2 Avg AC (100%) and OpenVLM OCRBench Official ocrbench_score_pct (88%)

19.1%
#2gemini-3-pro-preview

Strong on Vals SWE-bench overall_accuracy_pct (88%) and Vals LiveCodeBench overall_accuracy_pct (97%)

16.8%
#3gemini-2.5-pro

Strong on LEXam Leaderboard average_score_pct (89%) and Galileo Agent Leaderboard v2 Avg AC (59%)

16.2%
#6deepseek/deepseek-r1
15.7%
#7gpt-4.1-mini-20250414
15.4%
#8gpt-4o
15.4%
#11gpt-4o-2024-05-13
13.6%
#13gpt-5-2025-08-07
13.5%
#14google/gemini-3.1-pro-preview
13.2%
#15Grok-4-0709
13.0%
#17openai/gpt-5.4-2026-03-05
12.4%
#18claude-sonnet-4-20250514
12.2%
#19anthropic/claude-sonnet-4.6
12.1%
#20z-ai/glm-4.7
12.0%
#21anthropic/claude-opus-4-6-thinking
11.9%
#22claude-opus-4-5-20251101
11.7%
#23gemini-3-flash-preview
11.6%
#24minimax/minimax-m2.1
11.5%
#25gpt-5.2-2025-12-11
11.5%
#26gpt-5.1-2025-11-13
11.5%
#27anthropic/claude-opus-4-5-20251101-thinking
11.4%
#28Kimi K2 Thinking
11.3%
#31gpt-5-mini-2025-08-07
10.8%
#33gemini-2.5-flash
10.6%
#35kimi/kimi-k2.5-thinking
10.3%
#37anthropic/claude-sonnet-4-5-20250929-thinking
10.0%
#41gpt-4o-2024-08-06
9.7%
#47zai/glm-5-thinking
9.3%
#48qwen-2.5-72b-instruct
9.3%
#50xai-org/grok-4-fast-reasoning
9.0%

Compare Models

Model A leads by +2.3%

Shareable Link →

Model A

gpt-4.1-20250414

external/openai/gpt-4-1-20250414

19.1%

Rank #1

Confidence 30.1%23 evidence pts

Galileo Agent Leaderboard v2: Avg AC

Value 100.0% · Conf 100.0% · Weight 2.3%

galileo_agent_v2.avg_ac (Mar 12, 2026)

OpenVLM OCRBench Official: ocrbench_score_pct

Value 87.7% · Conf 100.0% · Weight 1.2%

openvlm_ocrbench_official.ocrbench_score_pct (Mar 12, 2026)

OpenVLM TextVQA Official: textvqa_score_pct

Value 76.8% · Conf 100.0% · Weight 1.0%

openvlm_textvqa_official.textvqa_score_pct (Mar 12, 2026)

OpenVLM MTVQA Official: mtvqa_score_pct

Value 92.4% · Conf 100.0% · Weight 0.9%

openvlm_mtvqa_official.mtvqa_score_pct (Mar 12, 2026)

Model B

gemini-3-pro-preview

external/google/gemini-3-pro-preview

16.8%

Rank #2

Confidence 20.8%22 evidence pts

Vals SWE-bench: overall_accuracy_pct

Value 87.5% · Conf 100.0% · Weight 1.4%

vals_swebench.overall_accuracy_pct (Mar 12, 2026)

Vals LiveCodeBench: overall_accuracy_pct

Value 97.1% · Conf 100.0% · Weight 1.4%

vals_lcb.overall_accuracy_pct (Mar 12, 2026)

LEXam Leaderboard: average_score_pct

Value 75.9% · Conf 100.0% · Weight 1.3%

lexam_leaderboard.average_score_pct (Mar 12, 2026)

Vals Terminal-Bench 2: overall_accuracy_pct

Value 81.0% · Conf 100.0% · Weight 1.2%

vals_terminal_bench_2.overall_accuracy_pct (Mar 12, 2026)

▶Ranking Diagnostics & Missing Models

Source Lift

Ranked

48

Sources

8

Quality

Insufficient

Vals LiveCodeBench

vals_lcb

37 rows

1.1% avg lift

Vals Tax Eval v2

vals_tax_eval_v2

37 rows

0.2% avg lift

Vals Legal Bench

vals_legal_bench

36 rows

0.3% avg lift

Vals GPQA

vals_gpqa

35 rows

0.2% avg lift

Missing Strong Models

x-ai/grok-3

external/x-ai/grok-3

Rank #35

12.1%

Thin evidence after weighting

anthropic/claude-opus-4-1-20250805

external/anthropic/claude-opus-4-1-20250805

Rank #48

10.8%

Thin evidence after weighting

deepseek-v3

external/deepseek-ai/deepseek-v3

Rank #66

8.8%

Thin evidence after weighting

Command A (03-2025)

external/cohere/command-a-03-2025

Rank #72

8.5%

Thin evidence after weighting
▶Taxonomy Details

Core Tasks

task.docstring_generationtask.write_report

Required Modes

none

Domains

domain.software_engineering

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