Papers › ORPO: Monolithic Preference Optimization without Reference Model

ORPO: Monolithic Preference Optimization without Reference Model

12 Mar 2024arXiv:2403.07691archive 2025-07-28

Jiwoo Hong, Noah Lee, James Thorne

While recent preference alignment algorithms for language models have demonstrated promising results, supervised fine-tuning (SFT) remains imperative for achieving successful convergence. In this paper, we study the crucial role of SFT within the context of preference alignment, emphasizing that a minor penalty for the disfavored generation style is sufficient for preference-aligned SFT. Building on this foundation, we introduce a straightforward and innovative reference model-free monolithic odds ratio preference optimization algorithm, ORPO, eliminating the necessity for an additional preference alignment phase. We demonstrate, both empirically and theoretically, that the odds ratio is a sensible choice for contrasting favored and disfavored styles during SFT across the diverse sizes from 125M to 7B. Specifically, fine-tuning Phi-2 (2.7B), Llama-2 (7B), and Mistral (7B) with ORPO on the UltraFeedback alone surpasses the performance of state-of-the-art language models with more than 7B and 13B parameters: achieving up to 12.20% on AlpacaEval_(2.0) (Figure 1), 66.19% on IFEval (instruction-level loose, Table 6), and 7.32 in MT-Bench (Figure 12). We release code and model checkpoints for Mistral-ORPO-α (7B) and Mistral-ORPO-β (7B).

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2403.07691")

Code

Syntology Ran 4 of 14 code samples harvested from 3 repositories linked to this paper; 10 have no recorded run. Of those that ran: 1 ran · honoured contract; 1 ran · our draft was wrong; 2 ran with no contract checked.

By repository: official repository: 3 samples from 1 repository, 2 ran; community (archive-listed): 11 samples from 2 repositories, 2 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

xfactlab/orpo officialmentioned in papermentioned on GitHubpytorchApache-2.0 report
internlm/xtuner mentioned on GitHubpytorch report
shibing624/medicalgpt mentioned on GitHubpytorchApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

14 samples harvested; 4 ran; 1 honoured the contract we drafted; 10 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
1ran · our draft was wrong
2ran
10unverified

Licence: 1 of the 14 samples is pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 repositories linked to this paper, official or community; each sample names its own and says which. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

dataset_split_selector xfactlab/orpo/src/utils.py official repository ran Apache-2.0 (permissive) · ea366da75b59ff77 · report
preprocess_logits_for_metrics xfactlab/orpo/src/utils.py official repository ran · honoured contract Apache-2.0 (permissive) · d2895bdf14e8f94f · report
default_args xfactlab/orpo/src/args.py official repository unverified Apache-2.0 (permissive) · 3e514a3233214a42 · report
ORPO OctopusMind/ORPO/orpo.py community (archive-listed) ran no licence file found · pointer only · 9da2f06d523b6949 · report
find_all_linear_names shibing624/medicalgpt/training/supervised_finetuning.py community (archive-listed) ran · our draft was wrong Apache-2.0 (permissive) · 7e7adc82abe38ce7 · report
add_extra_stop_words shibing624/medicalgpt/demo/openai_api.py community (archive-listed) unverified Apache-2.0 (permissive) · ac7e7d62416af65c · report
batch_generate_answer shibing624/medicalgpt/demo/inference.py community (archive-listed) unverified Apache-2.0 (permissive) · ce687202c5e103ad · report
compute_metrics shibing624/medicalgpt/training/reward_modeling.py community (archive-listed) unverified Apache-2.0 (permissive) · 69a87416659918fe · report
create_llm_client shibing624/medicalgpt/role_play_data/llm_client.py community (archive-listed) unverified Apache-2.0 (permissive) · afc0c3360d972be7 · report
find_all_linear_names shibing624/medicalgpt/training/reward_modeling.py community (archive-listed) unverified Apache-2.0 (permissive) · d640810687642a4f · report
generate shibing624/medicalgpt/role_play_data/roleplay_data_generate_doubao.py community (archive-listed) unverified Apache-2.0 (permissive) · 651d8039d4127f60 · report
stream_generate_answer shibing624/medicalgpt/demo/fastapi_server_demo.py community (archive-listed) unverified Apache-2.0 (permissive) · bf30ccc877226063 · report
stream_generate_answer shibing624/medicalgpt/demo/inference.py community (archive-listed) unverified Apache-2.0 (permissive) · b36a8f36f253e5c9 · report
trim_stop_words shibing624/medicalgpt/demo/openai_api.py community (archive-listed) unverified Apache-2.0 (permissive) · 469cf12b400c2c5a · report

Tasks

model

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

SFT

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections