Cross-Model LLM Code Review: Should you use Claude to review Codex or vice versa?
Not on the map. Why?
Abstract (arXiv)
Developers increasingly use two coding agents together: one writes a draft, and the other reviews it. However, it is not clear whether the pairing is worth its cost and time, or whether the order of the pairing matters. We run a controlled experiment on 116 recent hard and medium lcb tasks with Claude and Codex across six conditions to approximate a software practitioner's workflow: both solo baselines, both cross-model orderings, and both same-model orderings. The reviewer sees the problem and the writer's draft but cannot execute tests, which approximates a code review step. Claude review raises Codex drafts from 71.6% to 89.7% (pBH=.001); Codex self review raises them to 84.5% (pBH=.022). The reverse direction does not pay off: Codex reviewing Claude drafts drops the pass rate from 91.4% to 82.8% (pBH=.046), and Claude self review leaves the 91.4% baseline unchanged. Our evaluation indicates that the useful pairing is asymmetric: use Claude to review Codex, not the other way around.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 19 of its 37 references that are in the map. First posted on arXiv 22 Jul 2026.
Builds on
19 of 37 references.
References · 19
- 2021Evaluating Large Language Models Trained on Codecited by 8,053
- 2021Program Synthesis with Large Language Modelscited by 3,509
- 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,317
- 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 1,906
- 2022Competition-Level Code Generation with AlphaCodecited by 1,427
- 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 995
- 2023Teaching Large Language Models to Self-Debugcited by 963
- 2023ChatDev: Communicative Agents for Software Developmentcited by 893
- 2021Measuring Coding Challenge Competence With APPScited by 879
- 2023LEVER: Learning to Verify Language-to-Code Generation with Executioncited by 228
- 2023Is Self-Repair a Silver Bullet for Code Generation?cited by 191
- 2021Towards Automating Code Review Activitiescited by 54
7 more
- 2025CodeCoR: An LLM-Based Self-Reflective Multi-Agent Framework for Code Generationcited by 33
- 2026The Memory Curse: How Expanded Recall Erodes Cooperative Intent in LLM Agentscited by 10
- 2025DSPC: Dual-Stage Progressive Compression Framework for Efficient Long-Context Reasoningcited by 9
- 2026The Efficiency Frontier: A Unified Framework for Cost-Performance Optimization in LLM Context Managementcited by 6
- 2026Are LLMs Reliable Code Reviewers? Systematic Overcorrection in Requirement Conformance Judgementcited by 4
- 2026Accuracy, Stability, and Repeated-Run Reliability of Large Language Models on Deterministic Programming Taskscited by 4
- 2026SwiftBot: A Decentralized Platform for LLM-Powered Federated Robotic Task Executioncited by 1
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2026Reviewer Capability Governs Rejection Targeting, Not Repair Skill: Evidence from LLM Execute-Review-Revise Pipelines
- 2026TraceLab: Characterizing Coding Agent Workloads for LLM Serving
- 2026Review Arcade: On the Human Alignment and Gameability of LLM Reviews
- 2026Review Beats Planning: Dual-Model Interaction Patterns for Code Synthesis ↗
- 2026Cross-Model Cross-Language AI Coding Agent Performance: Accuracy and Speed of Parallel CLRS Algorithms
- 2026MathlibPR: Pull Request Merge-Readiness Benchmark for Formal Mathematical Libraries
- 2026More Code, Less Reuse: Investigating Code Quality and Reviewer Sentiment towards AI-generated Pull Requests
- 2026Agents Don't Paginate: First-Chunk Selection for LLM Tool Responses