reasoning benchmark
CRUXEval-Input-CoT
CRUXEval input prediction task with Chain of Thought (CoT) prompting. Part of the CRUXEval benchmark for code reasoning, understanding, and execution evaluation. Given a Python function and its expected output, the task is to predict the appropriate input using chain-of-thought reasoning. Consists of 800 Python functions (3-13 lines) designed to evaluate code comprehension and reasoning capabilities.
Updated Aug 7, 2026
Published models1
Registry coverage1
MetricScore
EvidenceB
CRUXEval-Input-CoT leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 56.5% | 100.0% | 1 | C |
CRUXEval-Input-CoT highlights
The top published results on this benchmark's own scale.
What is CRUXEval-Input-CoT?
Definition and scoring fields from the benchmark registry.
CRUXEval input prediction task with Chain of Thought (CoT) prompting. Part of the CRUXEval benchmark for code reasoning, understanding, and execution evaluation. Given a Python function and its expected output, the task is to predict the appropriate input using chain-of-thought reasoning. Consists of 800 Python functions (3-13 lines) designed to evaluate code comprehension and reasoning capabilities.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- CRUXEval-Input-CoT
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- cruxeval-input-cot|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about CRUXEval-Input-CoT.
Which model scores highest on CRUXEval-Input-CoT?
Qwen2.5-Coder 7B Instruct is currently ranked first with 56.5%.
What does CRUXEval-Input-CoT measure?
CRUXEval input prediction task with Chain of Thought (CoT) prompting. Part of the CRUXEval benchmark for code reasoning, understanding, and execution evaluation. Given a Python function and its expected output, the task is to predict the appropriate input using chain-of-thought reasoning. Consists of 800 Python functions (3-13 lines) designed to evaluate code comprehension and reasoning capabilities.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
1 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.