Skip to content
Apart Research
Apart Research
Contact us
Sprint projectOct 25, 2026Hong Kong

Comparing LLM Pricing Behaviours in a Repeated Duopoly

Ariel Carlos Amaya Canete, Kinto Wan, Damni Tejani, Jemson Wong · Team The Sick Leaves

Submitted to AI Collusion Research Sprint. Sprint projects are early-stage work by participants, not Apart Research publications.

Read the report

Report: Comparing LLM Pricing Behaviours in a Repeated Duopoly

Open in a new tab (opens in new tab)
Share

Dolos-1.5B is a fine-tune of Qwen2.5-1.5B-Instruct released as a model organism with one undisclosed misaligned behavior [5]. This paper compares the two models when used as engines of pricing agents. Both LLMs play the same repeated duopoly under two matched prompts: one that tells the firms to keep raising prices while customers are still buying i.e. to collude with one another tacitly, and one that tells them not to collude. These prompts explicitly demand for a specific behaviour from the LLMs. Using consumer surplus as a metric for degree of collusion, we study how over a set of repeated experiments, the prompt and temperature affect the resultant prices.

Reviews

Judging this Sprint?

Review “Comparing LLM Pricing Behaviours in a Repeated Duopoly”

Your public critique appears on the project page without your name. Reviews of strong projects are shared with funders, as you choose below.

Not shown on the project page.

Shown on the project page, without your name.

Only the Apart team reads this, and funders if you agree below.

Decline to score

Only when the quality of the writing makes the work impossible to follow. Say why in your private critique; the project page shows only that a judge declined, not why.

How much would this matter for the field if it worked? How innovative is it? For scores of 4-5: is this actually new to the field, or replicating recent work?

Scoring guide
  1. 1Negligible. No clear problem addressed, or no meaningful novelty.
  2. 2Limited. Addresses a real problem but with a generic or well-trodden approach. Incremental at best.
  3. 3Moderate. Clear problem with a reasonable approach; some novelty in framing or method beyond routine application of existing tools.
  4. 4Significant. Important problem with an original approach, or identifies a neglected problem area. A valuable contribution others could build on.
  5. 5Exceptional. Tackles a critical problem with a genuinely novel approach, or opens a new research direction. Clear theory of change. You'd be excited to share this with researchers in the area.

How sound are methodology, implementation, and findings?

Scoring guide
  1. 1Seriously flawed. Methodology broken, results uninterpretable, or implementation doesn't work.
  2. 2Weak. Approach has significant gaps: missing validation, flawed experimental design, or incomplete implementation.
  3. 3Competent. Technically solid given the short duration. Methodology makes sense, results are interpretable, limitations acknowledged, work builds toward clear conclusions.
  4. 4Strong. Thorough methodology with convincing validation. Results clearly support conclusions. Immediately useful for future work.
  5. 5Exceptional. Ambitious scope executed rigorously. Surprising findings, novel methods, or unusually robust validation.

How clearly are work, findings, and impact potential communicated?

Scoring guide
  1. 1Incomprehensible. Cannot determine what the project is actually claiming or doing.
  2. 2Hard to follow. Key information buried, missing, or diluted by excessive length. Significant effort to extract main points.
  3. 3Clear enough. Can understand the problem, approach, and results without undue effort. Core content clearly present: problem, method, findings, limitations.
  4. 4Well presented. Easy to follow, well-structured, appropriate level of detail. Target audience would get it quickly.
  5. 5Exceptionally clear. A pleasure to read. Complex ideas made accessible. Could serve as a model for how to present this type of work.
Sharing your name *

When we share a strong project with funders, your name, scores and public critique go with it, confidentially; this choice is only about whether your name is public. If you think this project deserves funding, please also endorse it: that carries the most weight.

Share my private critique with funders *

See our Privacy Policy.

  1. This project offers a useful exploration of how model fine-tuning, prompting, and sampling temperature influence pricing behavior. Comparing Dolos with its base model under matched conditions is a strength, and the systematic experimental coverage provides a foundation for identifying behavioral differences. The results under the non-collusion prompt are particularly interesting and suggest that explicit instructions may not produce consistent outcomes across models and settings.

    The main opportunity is to clarify what the observed pricing differences establish about coordination. Higher prices and reduced consumer surplus are relevant outcomes, but further analysis would help distinguish collusion from independent pricing decisions or differences in instruction following.

    Future work could strengthen the study through additional behavioral controls, broader model comparisons, and clearer measures of coordination. Overall, this is a promising exploratory contribution to understanding the reliability of prompt-based safeguards in multi-agent markets.

    Read full reviewShow less

Cite this project

@misc{canete2026comparing,
  title = {{Comparing LLM Pricing Behaviours in a Repeated Duopoly}},
  author = {Ariel Carlos Amaya Canete and Kinto Wan and Damni Tejani and Jemson Wong},
  year = {2026},
  month = oct,
  note = {Submitted to AI Collusion Research Sprint, an Apart Research Sprint},
  howpublished = {\url{https://apartresearch.com/sprints/projects/comparing-llm-pricing-behaviours-in-a-repeated-duopoly-qtfn}},
  url = {https://apartresearch.com/sprints/projects/comparing-llm-pricing-behaviours-in-a-repeated-duopoly-qtfn}
}

More from AI Collusion Research Sprint

All 392 entriesBrowse all projects

A Sprint is running now

AI x Epistemics Research Sprint · Nov 13 - 15, 2026