Papers › Archon: An Architecture Search Framework for Inference-Time Techniques

Archon: An Architecture Search Framework for Inference-Time Techniques

23 Sep 2024arXiv:2409.15254archive 2025-07-28

Jon Saad-Falcon, Adrian Gamarra Lafuente, Shlok Natarajan, Nahum Maru, Hristo Todorov, Etash Guha, E. Kelly Buchanan, Mayee Chen, Neel Guha, Christopher Ré, Azalia Mirhoseini

Inference-time techniques are emerging as highly effective tools to enhance large language model (LLM) capabilities. However, best practices for developing systems that combine these techniques remain underdeveloped due to our limited understanding of the utility of individual inference-time techniques and the interactions between them. Additionally, efficiently and automatically searching the space of model choices, inference-time techniques, and their compositions is challenging due to the large design space. To address these challenges, we introduce Archon, a modular framework for selecting, combining, and stacking layers of inference-time techniques to construct optimized LLM systems for target benchmarks. Rather than relying on a single LLM called once, we leverage a diverse set of LLMs and inference-time techniques, creating LLM systems greater than the sum of their parts. Archon defines an extensible design space, encompassing techniques such as generation ensembling, repeated sampling, ranking, fusion, critiquing, verification, and unit testing. It transforms the problem of building LLM systems into a hyperparameter optimization objective. Given the available LLMs, inference-time techniques, and compute budget, Archon utilizes hyperparameter search techniques to discover optimized architectures for target benchmark(s). We evaluate Archon architectures across a range of instruction-following, reasoning, and coding benchmarks, including MT-Bench, Arena-Hard-Auto, AlpacaEval 2.0, MixEval, MixEval Hard, MATH, and CodeContests. Archon architectures outperform frontier models, such as GPT-4o and Claude 3.5 Sonnet, on these benchmarks, achieving an average accuracy increase of 15.1 percentage points by using all available LLMs. We make our code and datasets available publicly on Github: https://github.com/ScalingIntelligence/Archon.

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 16 code samples harvested from 1 repository linked to this paper: 14 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; 9 where Syntology's instrument failed) · 2 unverified. Where Syntology's instrument failed: 9 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 16 samples from 1 repository, 14 ran (5 with no instrument failure; 9 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 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

scalingintelligence/archon officialmentioned in papermentioned on GitHubApache-2.0Syntology: 16 samples, 14 ran (9 instrument-failed) · 2 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

16 samples harvested: 14 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; 9 where Syntology's instrument failed) · 2 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
9ran · our draft was wrong
4ran
2unverified

Licence: 0 of the 16 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 scalingintelligence/archon. “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.

scalingintelligence/archon official repository 16 samples, 14 ran (9 instrument-failed) · 2 unverified
append_json src/archon/itas_algorithms/itas_algorithm.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 0ff4a203cd88039c · report fixture: signature file_path:str/0/str · returned str · level recorded 2026-09-24
clean_messages src/archon/completions/utils.py ran · our draft was wrong Apache-2.0 (permissive) · 107600cb9ca038b0 · report fixture: signature messages:list/1/dict · returned list · level recorded 2026-09-24
compute_mle_elo src/archon/benchmarks/arena_hard_auto/show_arena_hard_auto_result.py ran Apache-2.0 (permissive) · e699241c4fab3815 · report fixture: signature df:DataFrame/0/?, SCALE:int/0/int, BASE:int/0/int, INIT_RATING:int/0/int, baseline_model:str/0/str · returned Series · level recorded 2026-09-24
count_correct_yaml_files src/archon/benchmarks/code_contests/show_code_results.py ran · honoured contract Apache-2.0 (permissive) · e4f095487017d03d · report fixture: signature eval_path:str/0/str · returned tuple · level recorded 2026-09-24
format_prompt in 45 papers src/archon/completions/utils.py ran · our draft was wrong Apache-2.0 (permissive) · a3621beb3eb4d82a · report fixture: signature messages:list/1/dict · returned str · level recorded 2026-09-24
generate_pairwise_config src/archon/benchmarks/arena_hard_auto/gen_judgment.py ran Apache-2.0 (permissive) · 96970be6c2e04513 · report fixture: signature baseline:str/0/str, judge:str/0/str, model_list:list/1/str, temperature:float/0/float · returned dict · level recorded 2026-09-24
get_bootstrap_result src/archon/benchmarks/arena_hard_auto/show_arena_hard_auto_result.py ran Apache-2.0 (permissive) · 4a901a353cc86998 · report fixture: signature battles:DataFrame/0/?, func_compute_elo:callable/0/?, num_round:int/0/int, baseline_model:str/0/str · returned DataFrame · level recorded 2026-09-24
get_endpoint src/archon/benchmarks/arena_hard_auto/arena_hard_auto_utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · c8136157d71c093b · report fixture: signature endpoint_list:list/1/dict · returned dict · level recorded 2026-09-24
load_config in 389 papers src/archon/completions/utils.py ran · our draft was wrong Apache-2.0 (permissive) · a92705181ff634e3 · report fixture: signature config_path:str/0/str · returned dict · level recorded 2026-09-24
load_model_answers in 5 papers src/archon/benchmarks/arena_hard_auto/arena_hard_auto_utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 28bbb1e46eb5f17c · report fixture: signature answer_dir:str/0/str · returned dict · level recorded 2026-09-24
load_questions in 17 papers src/archon/benchmarks/arena_hard_auto/arena_hard_auto_utils.py ran · our draft was wrong Apache-2.0 (permissive) · 47892c5631a199d4 · report fixture: signature question_file:str/0/str · returned list · level recorded 2026-09-24
load_search_config src/archon/itas_algorithms/itas_algorithm.py ran · our draft was wrong Apache-2.0 (permissive) · 047d950b82412f4f · report fixture: signature file_path:str/0/str · returned dict · level recorded 2026-09-24
parse_model_name in 11 papers src/archon/itas_algorithms/power_ranker.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · cec182ddf2df6b1e · report fixture: signature model_path:str/0/str · returned str · level recorded 2026-09-24
preety_print_two_ratings src/archon/benchmarks/arena_hard_auto/show_arena_hard_auto_result.py ran Apache-2.0 (permissive) · 2a7dcca7dcf80307 · report fixture: signature ratings_1:dict/0/?, ratings_2:dict/0/?, column_names:list/1/str · returned DataFrame · level recorded 2026-09-24
get_score in 56 papers src/archon/benchmarks/arena_hard_auto/gen_judgment.py unverified Apache-2.0 (permissive) · b2d83f0b138fe334 · report
load_benchmark in 6 papers src/archon/benchmarks/benchmarks.py unverified Apache-2.0 (permissive) · 4a737f68f1c25eef · report

Tasks

Hyperparameter OptimizationInstruction FollowingLarge Language ModelMath

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

SET

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