Papers › OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data

OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Data

2 Oct 2024arXiv:2410.01560archive 2025-07-28

Shubham Toshniwal, Wei Du, Ivan Moshkov, Branislav Kisacanin, Alexan Ayrapetyan, Igor Gitman

Mathematical reasoning continues to be a critical challenge in large language model (LLM) development with significant interest. However, most of the cutting-edge progress in mathematical reasoning with LLMs has become closed-source due to lack of access to training data. This lack of data access limits researchers from understanding the impact of different choices for synthesizing and utilizing the data. With the goal of creating a high-quality finetuning (SFT) dataset for math reasoning, we conduct careful ablation experiments on data synthesis using the recently released Llama3.1 family of models. Our experiments show that: (a) solution format matters, with excessively verbose solutions proving detrimental to SFT performance, (b) data generated by a strong teacher outperforms equally-sized data generated by a weak student model, (c) SFT is robust to low-quality solutions, allowing for imprecise data filtering, and (d) question diversity is crucial for achieving data scaling gains. Based on these insights, we create the OpenMathInstruct-2 dataset, which consists of 14M question-solution pairs (≈ 600K unique questions), making it nearly eight times larger than the previous largest open-source math reasoning dataset. Finetuning the Llama-3.1-8B-Base using OpenMathInstruct-2 outperforms Llama3.1-8B-Instruct on MATH by an absolute 15.9% (51.9% → 67.8%). Finally, to accelerate the open-source efforts, we release the code, the finetuned models, and the OpenMathInstruct-2 dataset under a commercially permissive license.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 15 code samples harvested from 1 repository linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 7 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 15 samples from 1 repository, 12 ran (5 with no instrument failure; 7 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

NVIDIA/NeMo-Skills mentioned on GitHubApache-2.0Syntology: 15 samples, 12 ran (7 instrument-failed) · 3 unverified 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

15 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 7 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
7ran · our draft was wrong
4ran
3unverified

Licence: 0 of the 15 samples are 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 NVIDIA/NeMo-Skills. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

NVIDIA/NeMo-Skills community (archive-listed) 15 samples, 12 ran (7 instrument-failed) · 3 unverified
calculate_chunk_indices nemo_skills/file_utils.py ran · honoured contract Apache-2.0 (permissive) · 2e7aafbc230cb972 · report fixture: signature num_samples:int/0/int, num_chunks:int/0/int, chunk_id:int/0/int · returned tuple · level recorded 2026-09-25
determine_proof_status nemo_skills/code_execution/proof_utils.py ran · our draft was wrong Apache-2.0 (permissive) · dfb61cd2ac2b564a · report fixture: signature compiler_output:dict/0/? · returned str · level recorded 2026-09-25
extract_code_output nemo_skills/code_execution/utils.py ran fingerprinted Apache-2.0 (permissive) · b4bbfbf06bb92de6 · report fixture: signature generation:str/0/str, code_output_begin:str/0/str, code_output_end:str/0/str, extract_all:bool/0/bool · returned str · level recorded 2026-09-25
extract_code_to_execute nemo_skills/code_execution/utils.py ran fingerprinted Apache-2.0 (permissive) · b75a1053974f21ce · report fixture: signature generation:str/0/str, code_begin:str/0/str, code_end:str/0/str, extract_all:bool/0/bool · returned str · level recorded 2026-09-25
extract_proof_only nemo_skills/code_execution/proof_utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · df5a88c92c8f57ce · report fixture: signature lean_code:str/0/str · returned str · level recorded 2026-09-25
format_code_output nemo_skills/code_execution/utils.py ran Apache-2.0 (permissive) · b2186409cb2b7dfb · report fixture: signature execution_dict:dict/0/?, code_output_begin:str/0/str, code_output_end:str/0/str, code_output_format:str/0/str, remaining_code_executions:int/0/int · returned str · level recorded 2026-09-25
get_logger_name nemo_skills/utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 364d75dfb259e939 · report fixture: signature file:str/0/str · returned str · level recorded 2026-09-25
import_score_module nemo_skills/evaluation/compute_group_score.py ran · our draft was wrong Apache-2.0 (permissive) · 28ecc1878ccced7c · report fixture: signature score_module:str/0/str · returned module · level recorded 2026-09-25
init_wandb in 6 papers nemo_skills/utils.py ran Apache-2.0 (permissive) · a2afdfc356abe1d1 · report fixture: signature project:str/0/str, name:str/0/str, exp_dir:str/0/str, verbose:bool/0/bool · returned bool · level recorded 2026-09-25
jload in 59 papers nemo_skills/file_utils.py ran · our draft was wrong Apache-2.0 (permissive) · 8c48543062658550 · report fixture: signature filepath:str/0/str, mode:str/0/str, verbose:bool/0/bool · returned list · level recorded 2026-09-25
load_metric_files nemo_skills/evaluation/compute_group_score.py ran · our draft was wrong Apache-2.0 (permissive) · 6f55869f2815d973 · report fixture: signature metric_files:list/1/str · returned dict · level recorded 2026-09-25
setup_logging in 49 papers nemo_skills/utils.py ran · our draft was wrong Apache-2.0 (permissive) · 4261064eb4ec7756 · report fixture: signature disable_hydra_logs:bool/0/bool, log_level:int/0/int, use_rich:bool/0/bool · returned Logger · level recorded 2026-09-25
count_newlines nemo_skills/file_utils.py unverified Apache-2.0 (permissive) · 5b3c40e3214faf87 · report
load_config in 389 papers nemo_skills/conversion/hf_to_nemo_llama.py unverified Apache-2.0 (permissive) · e74354657e62b51d · report
load_config in 389 papers nemo_skills/conversion/hf_to_nemo_qwen.py unverified Apache-2.0 (permissive) · de5cce096268a65e · report

Tasks

Arithmetic ReasoningLarge Language ModelMathMath Word Problem SolvingMathematical Reasoning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Arithmetic Reasoning GSM8K OpenMath2-Llama3.1-70B (majority@256) Accuracy 96.0 #5 of 164 Archive leaderboard report
Arithmetic Reasoning GSM8K OpenMath2-Llama3.1-70B Accuracy 94.9 #9 of 164 Archive leaderboard report
Arithmetic Reasoning GSM8K OpenMath2-Llama3.1-8B (majority@256) Accuracy 94.1 #12 of 164 Archive leaderboard report
Arithmetic Reasoning GSM8K OpenMath2-Llama3.1-8B Accuracy 91.7 #17 of 164 Archive leaderboard report
Math Word Problem Solving MATH OpenMath2-Llama3.1-70B (majority@256) Accuracy 79.6 #10 of 135 Archive leaderboard report
Math Word Problem Solving MATH OpenMath2-Llama3.1-8B (majority@256) Accuracy 76.1 #11 of 135 Archive leaderboard report
Math Word Problem Solving MATH OpenMath2-Llama3.1-70B Accuracy 71.9 #15 of 135 Archive leaderboard report
Math Word Problem Solving MATH OpenMath2-Llama3.1-8B Accuracy 67.8 #19 of 135 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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