Papers › LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model

LLaMA-Adapter V2: Parameter-Efficient Visual Instruction Model

28 Apr 2023arXiv:2304.15010archive 2025-07-28

Peng Gao, Jiaming Han, Renrui Zhang, Ziyi Lin, Shijie Geng, Aojun Zhou, Wei zhang, Pan Lu, Conghui He, Xiangyu Yue, Hongsheng Li, Yu Qiao

How to efficiently transform large language models (LLMs) into instruction followers is recently a popular research direction, while training LLM for multi-modal reasoning remains less explored. Although the recent LLaMA-Adapter demonstrates the potential to handle visual inputs with LLMs, it still cannot generalize well to open-ended visual instructions and lags behind GPT-4. In this paper, we present LLaMA-Adapter V2, a parameter-efficient visual instruction model. Specifically, we first augment LLaMA-Adapter by unlocking more learnable parameters (e.g., norm, bias and scale), which distribute the instruction-following ability across the entire LLaMA model besides adapters. Secondly, we propose an early fusion strategy to feed visual tokens only into the early LLM layers, contributing to better visual knowledge incorporation. Thirdly, a joint training paradigm of image-text pairs and instruction-following data is introduced by optimizing disjoint groups of learnable parameters. This strategy effectively alleviates the interference between the two tasks of image-text alignment and instruction following and achieves strong multi-modal reasoning with only a small-scale image-text and instruction dataset. During inference, we incorporate additional expert models (e.g. captioning/OCR systems) into LLaMA-Adapter to further enhance its image understanding capability without incurring training costs. Compared to the original LLaMA-Adapter, our LLaMA-Adapter V2 can perform open-ended multi-modal instructions by merely introducing 14M parameters over LLaMA. The newly designed framework also exhibits stronger language-only instruction-following capabilities and even excels in chat interactions. Our code and models are available at https://github.com/ZrrSkywalker/LLaMA-Adapter.

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

The archive's official repository (zrrskywalker/llama-adapter) was not harvested; the 1 sample below comes from 1 community repository.

Syntology 1 code sample harvested from 1 repository linked to this paper: 0 ran · 1 unverified.

By repository: community (archive-listed): 1 sample from 1 repository, 0 ran (0 with no instrument failure; 0 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.

zrrskywalker/llama-adapter officialmentioned in papermentioned on GitHubSyntology: not harvested report
opengvlab/llama-adapter mentioned on GitHubpytorchGPL-3.0Syntology: 1 sample, 0 ran · 1 unverified report
Mind23-2/MindCode-140 mindsporeSyntology: not harvested 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

1 sample harvested: 0 ran · 1 unverified. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1unverified

Licence: 1 of the 1 sample 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 opengvlab/llama-adapter. 1 sample whose link to this paper records no repository is listed under one of this paper's own repositories, because identical code was harvested from it for another paper. “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 “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.

opengvlab/llama-adapter community (archive-listed) 1 sample, 0 ran · 1 unverified
imagebind_huge in 7 papers imagebind_LLM/ImageBind/models/imagebind_model.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own unverified licence of this copy not recorded · 71c517c59227454e · report

Tasks

Instruction FollowingOptical Character Recognition (OCR)Video Question AnsweringVideo-based Generative Performance BenchmarkingVideo-based Generative Performance Benchmarking (Consistency)Video-based Generative Performance Benchmarking (Contextual Understanding)Video-based Generative Performance Benchmarking (Correctness of Information)Video-based Generative Performance Benchmarking (Detail Orientation))Video-based Generative Performance Benchmarking (Temporal Understanding)Visual Question AnsweringVisual Question Answering (VQA)Zero-Shot Video Question Answermodel

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Question Answering ActivityNet-QA LLaMA Adapter V2 Accuracy 34.2 #30 of 36 Archive leaderboard report
Video Question Answering ActivityNet-QA LLaMA Adapter V2 Confidence score 2.7 #30 of 36 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct LLaMA Adapter Consistency 2.15 #22 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct LLaMA Adapter Contextual Understanding 2.30 #22 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct LLaMA Adapter Correctness of Information 2.03 #22 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct LLaMA Adapter Detail Orientation 2.32 #22 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct LLaMA Adapter Temporal Understanding 1.98 #22 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking VideoInstruct LLaMA Adapter mean 2.16 #22 of 23 Archive leaderboard report
Video-based Generative Performance Benchmarking (Consistency) VideoInstruct LLaMA Adapter gpt-score 2.15 #17 of 18 Archive leaderboard report
Video-based Generative Performance Benchmarking (Contextual Understanding) VideoInstruct LLaMA Adapter gpt-score 2.30 #17 of 18 Archive leaderboard report
Video-based Generative Performance Benchmarking (Correctness of Information) VideoInstruct LLaMA Adapter gpt-score 2.03 #17 of 18 Archive leaderboard report
Video-based Generative Performance Benchmarking (Detail Orientation)) VideoInstruct LLaMA Adapter gpt-score 2.32 #17 of 18 Archive leaderboard report
Video-based Generative Performance Benchmarking (Temporal Understanding) VideoInstruct LLaMA Adapter gpt-score 1.98 #16 of 18 Archive leaderboard report
Visual Question Answering MM-Vet LLaMA-Adapter v2-7B GPT-4 score 31.4±0.1 #194 of 231 Archive leaderboard report
Visual Question Answering MM-Vet LLaMA-Adapter v2-7B Params 7B #194 of 231 Archive leaderboard report
Visual Question Answering (VQA) InfiMM-Eval LLaMA-Adapter V2 Abductive 46.12 #6 of 14 Archive leaderboard report
Visual Question Answering (VQA) InfiMM-Eval LLaMA-Adapter V2 Analogical 22.08 #6 of 14 Archive leaderboard report
Visual Question Answering (VQA) InfiMM-Eval LLaMA-Adapter V2 Deductive 28.7 #6 of 14 Archive leaderboard report
Visual Question Answering (VQA) InfiMM-Eval LLaMA-Adapter V2 Overall score 30.46 #6 of 14 Archive leaderboard report
Visual Question Answering (VQA) InfiMM-Eval LLaMA-Adapter V2 Params 7B #6 of 14 Archive leaderboard report
Zero-Shot Video Question Answer ActivityNet-QA LLaMA Adapter Accuracy 34.2 #25 of 28 Archive leaderboard report
Zero-Shot Video Question Answer ActivityNet-QA LLaMA Adapter Confidence Score 2.7 #25 of 28 Archive leaderboard report
Zero-Shot Video Question Answer MSRVTT-QA LLaMA Adapter-7B Accuracy 43.8 #29 of 30 Archive leaderboard report
Zero-Shot Video Question Answer MSRVTT-QA LLaMA Adapter-7B Confidence Score 2.7 #29 of 30 Archive leaderboard report
Zero-Shot Video Question Answer MSVD-QA LLaMA Adapter-7B Accuracy 54.9 #26 of 28 Archive leaderboard report
Zero-Shot Video Question Answer MSVD-QA LLaMA Adapter-7B Confidence Score 3.1 #26 of 28 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

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGPT-4Label SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

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