Papers › Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters

6 Aug 2024arXiv:2408.03314archive 2025-07-28

Charlie Snell, Jaehoon Lee, Kelvin Xu, Aviral Kumar

Enabling LLMs to improve their outputs by using more test-time computation is a critical step towards building generally self-improving agents that can operate on open-ended natural language. In this paper, we study the scaling of inference-time computation in LLMs, with a focus on answering the question: if an LLM is allowed to use a fixed but non-trivial amount of inference-time compute, how much can it improve its performance on a challenging prompt? Answering this question has implications not only on the achievable performance of LLMs, but also on the future of LLM pretraining and how one should tradeoff inference-time and pre-training compute. Despite its importance, little research attempted to understand the scaling behaviors of various test-time inference methods. Moreover, current work largely provides negative results for a number of these strategies. In this work, we analyze two primary mechanisms to scale test-time computation: (1) searching against dense, process-based verifier reward models; and (2) updating the model's distribution over a response adaptively, given the prompt at test time. We find that in both cases, the effectiveness of different approaches to scaling test-time compute critically varies depending on the difficulty of the prompt. This observation motivates applying a "compute-optimal" scaling strategy, which acts to most effectively allocate test-time compute adaptively per prompt. Using this compute-optimal strategy, we can improve the efficiency of test-time compute scaling by more than 4x compared to a best-of-N baseline. Additionally, in a FLOPs-matched evaluation, we find that on problems where a smaller base model attains somewhat non-trivial success rates, test-time compute can be used to outperform a 14x larger model.

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 9 code samples harvested from 1 repository linked to this paper: 6 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 9 samples from 1 repository, 6 ran (5 with no instrument failure; 1 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.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

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

huggingface/search-and-learn mentioned on GitHubpytorchApache-2.0Syntology: 9 samples, 6 ran (1 instrument-failed) · 3 unverified report
zarahall/fairness-prms mentioned on GitHubSyntology: not harvested report
codelion/optillm pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Self-Consistency Improves Chain of Thought Reasoning in Language Models (arXiv:2203.11171; that paper's own run record: 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result), Tree of Thoughts: Deliberate Problem Solving with Large Language Models (arXiv:2305.10601; that paper's own run record: 19 ran (of which 6 constructed an object rather than computing a result; 17 with no instrument failure: 0 honoured, 0 violated, 17 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified) and Unsupervised Evaluation of Code LLMs with Round-Trip Correctness (arXiv:2402.08699; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

9 samples harvested: 6 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 1 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 · our draft was wrong
5ran
3unverified

Licence: 8 of the 9 samples are pointer only, meaning Syntology does not serve that copy's text. For 8 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from huggingface/search-and-learn. “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.

huggingface/search-and-learn community (archive-listed) 9 samples, 6 ran (1 instrument-failed) · 3 unverified
aggregate_scores src/sal/utils/score.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · e2fa8778aaa14477 · report fixture: signature scores:list/1/float, agg_strategy:str/0/str · returned float · level recorded 2026-09-25
build_conv in 7 papers src/sal/search/utils.py ran fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · fcb58cf3620ba3fa · report fixture: signature prompt:str/0/str, response:str/0/str, system_prompt:str/0/str · returned list · level recorded 2026-09-25
extract_multi_choice_answer in 13 papers src/sal/utils/qwen_math_parser.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · ed2f4689128f9885 · report fixture: signature pred_str:str/0/str · returned str · level recorded 2026-09-25
last src/sal/search/utils.py ran fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · 0a2729199ea72cd2 · report fixture: signature x:list/1/int · returned int · level recorded 2026-09-25
list_mean src/sal/search/utils.py ran fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · e1583186b85bad63 · report fixture: signature x:list/1/float · returned float64 · level recorded 2026-09-25
subsample_completions src/sal/utils/math.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 940f44544351a853 · report fixture: signature x:dict/0/?, n:int/0/int · returned dict · level recorded 2026-09-25
batched_math_shepherd_inference src/sal/models/reward_models.py unverified Apache-2.0 (permissive) · 0f5e24b2cb4f615b · report
extract_completion_answers src/sal/utils/math.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 3c4ade317b7b432b · report
get_dataset_revisions src/sal/utils/hub.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 07e55a87bf057096 · report

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

BASEFocus

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