The Best Programming Language for Tokenmaxxing: An Investigation of Coding Agent Behavior Across Programming Languages

• URL: http://arxiv.org/abs/2607.22807v1
• Date: Fri, 24 Jul 2026 16:41:59 GMT
• Title: The Best Programming Language for Tokenmaxxing: An Investigation of Coding Agent Behavior Across Programming Languages
• Abstract summary: We evaluate five recent models on programming problems in Python, Java, Rust, and OCaml.<n>We show that there can be stark variation in token consumption that is consistent across models.
• Score: 13.509521852529728
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Although coding agents are now very effective in a variety of programming languages, this paper first shows that the cost (in tokens) can very significantly by programming language. We evaluate five recent models on programming problems in Python, Java, Rust, and OCaml. We carefully control for problem difficulty, and show that there can be stark variation in token consumption that is consistent across models. To understand why, we analyze both the structure and content of agent trajectories. First, we re-execute every intermediate solution and abstract each trajectory as a sequence of test-outcome vectors, then label the work between successive solutions. This reveals agents repeatedly producing noncompiling solutions in unfamiliar languages and revising solutions that already pass. Second, we analyze trajectory text, finding that agents plan solutions in code comments, distrust the provided tests in favor of inputs they invent, and sidestep unfamiliar target languages by prototyping in Python. Our results show that by-language token efficiency is a metric that should be considered when benchmarking and developing multilingual agents, and, for the tokenmaxxer, a guide to the most expensive language to work in.

Related papers

• Large Language Models for Multilingual Code Intelligence: A Survey [15.859504186659825]
  This survey focuses on two key tasks: multilingual code generation from shared natural-language requirements, and multilingual code translation that preserves semantics across languages.<n>It reviews representative methods, benchmarks, and evaluation metrics, and highlights challenges and opportunities for trustworthy cross-language generalization.
  arXiv  Detail & Related papers  (2026-04-27T20:20:26Z)
• Evaluating Robustness of Large Language Models Against Multilingual Typographical Errors [45.37878669586302]
  Large language models (LLMs) are increasingly deployed in multilingual, real-world applications with user inputs.<n>Most benchmarks assume clean input, leaving the robustness of LLMs to typos largely underexplored.<n>We introduce MulTypo, a multilingual typo generation algorithm that simulates human-like errors based on language-specific keyboard layouts and typing behavior.
  arXiv  Detail & Related papers  (2025-10-10T16:49:12Z)
• Isolating Language-Coding from Problem-Solving: Benchmarking LLMs with PseudoEval [7.33924106492889]
  Existing code generation benchmarks are designed to study Large Language Models' end-to-end performance.<n>We construct PseudoEval, a multilingual code generation benchmark that provides a solution written in pseudocode as input.<n>Our study indicates that problem-solving capability may transfer across programming languages, while language-coding needs more language-specific effort.
  arXiv  Detail & Related papers  (2025-02-26T14:08:17Z)
• CRUXEval-X: A Benchmark for Multilingual Code Reasoning, Understanding and Execution [50.1875460416205]
  The CRUXEVAL-X code reasoning benchmark contains 19 programming languages.<n>It comprises at least 600 subjects for each language, along with 19K content-consistent tests in total.<n>Even a model trained solely on Python can achieve at most 34.4% Pass@1 in other languages.
  arXiv  Detail & Related papers  (2024-08-23T11:43:00Z)
• Multilingual Contrastive Decoding via Language-Agnostic Layers Skipping [60.458273797431836]
  Decoding by contrasting layers (DoLa) is designed to improve the generation quality of large language models. We find that this approach does not work well on non-English tasks. Inspired by previous interpretability work on language transition during the model's forward pass, we propose an improved contrastive decoding algorithm.
  arXiv  Detail & Related papers  (2024-07-15T15:14:01Z)
• Teaching Large Language Models to Self-Debug [62.424077000154945]
  Large language models (LLMs) have achieved impressive performance on code generation. We propose Self- Debugging, which teaches a large language model to debug its predicted program via few-shot demonstrations.
  arXiv  Detail & Related papers  (2023-04-11T10:43:43Z)
• xCodeEval: A Large Scale Multilingual Multitask Benchmark for Code Understanding, Generation, Translation and Retrieval [32.60391966381949]
  We introduce xCodeEval, the largest executable multilingual multitask benchmark to date. It features a total of $7$ tasks involving code understanding, generation, translation and retrieval. xCodeEval adopts an execution-based evaluation and offers a multilingual code execution engine, ExecEval.
  arXiv  Detail & Related papers  (2023-03-06T10:08:51Z)
• Language Models Can Teach Themselves to Program Better [4.627023679353507]
  Recent Language Models (LMs) achieve breakthrough performance in code generation when trained on human-authored problems. We show that it is possible for an LM to synthesize programming problems and solutions, which are filtered for correctness by a Python interpreter. The LM's performance is then seen to improve when it is fine-tuned on its own synthetic problems and verified solutions.
  arXiv  Detail & Related papers  (2022-07-29T06:43:28Z)
• AVATAR: A Parallel Corpus for Java-Python Program Translation [77.86173793901139]
  Program translation refers to migrating source code from one language to another. We present AVATAR, a collection of 9,515 programming problems and their solutions written in two popular languages, Java and Python.
  arXiv  Detail & Related papers  (2021-08-26T05:44:20Z)
• Revisiting Language Encoding in Learning Multilingual Representations [70.01772581545103]
  We propose a new approach called Cross-lingual Language Projection (XLP) to replace language embedding. XLP projects the word embeddings into language-specific semantic space, and then the projected embeddings will be fed into the Transformer model. Experiments show that XLP can freely and significantly boost the model performance on extensive multilingual benchmark datasets.
  arXiv  Detail & Related papers  (2021-02-16T18:47:10Z)
• X-FACTR: Multilingual Factual Knowledge Retrieval from Pretrained Language Models [103.75890012041366]
  Language models (LMs) have proven surprisingly successful at capturing factual knowledge. However, studies on LMs' factual representation ability have almost invariably been performed on English. We create a benchmark of cloze-style probes for 23 typologically diverse languages.
  arXiv  Detail & Related papers  (2020-10-13T05:29:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.