TraceEval - Python: leaderboard

TraceEval (2026) asks LLMs to list the caller-callee edges of a program call graph; this board covers the 769 Python test programs. Edge-level F1 in percent against edges observed by executing the program, zero-shot at temperature 0; higher is better.

Metric: F1 (%) on the 769 Python test programs, edge-level F1 of the predicted call graph against execution-verified edges, zero-shot at temperature 0; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 10 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 4.679.1#113
2Claude Sonnet 4.671.3#173
3GPT-5.466.6#135
4DeepSeek V3.263.1#419
5GPT-5.4 Mini59.7#406
6Qwen 2.5 Coder 32B Instruct57.1#1547
7Llama 3.3 70B Instruct53.3#1385
8Gemini 3.1 Pro (Preview)49.5#99
9Qwen 2.5 Coder 7B Instruct27.6#3914
10Qwen2.5-Coder-1.5B-Instruct16.1#8221

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=traceeval-python · How It Works · Data refreshed daily, snapshot 2026-11-21.