Try Kling Image V4 Long for Page Mockups from Kling AI →
Models
Agents
Workflows
Studio
PricingBlogDocs
ExploreDiscover models by categoryDiscoverModels and effects built by WiroBrowse All ModelsBrowse the complete catalogSee FavoritesSign in to view saved models
Generative Media AgentCreate and edit media by chattingWorkflow AgentBuild visual workflows with Agent
OverviewThe platform at a glanceLearnSkills, knowledge, guardrailsAnatomyWhat makes agents reasonBuild Your AgentPick connectors and skills, choose a plan, deploy
Pre-built AgentsBrowse the catalog
Agent Usecases
Ad Campaign ManagerApp Event ManagerApp Review RepliesBarber BookingCustomer Win-BackEcommerce ListingsRestaurant Reviews
Sign InStart Building
Explore models/
Video GenerationActive

VideoLLaMA3 2B

bydamo-nlp-sg

DAMO-NLP-SG's 2B VideoLLaMA 3 model watches a video clip and answers in English text. It skips image patches that stay the same from frame to frame, so static shots use fewer tokens.

Video to Text
Model ID
VideoLLaMA3-2B
Provider
damo-nlp-sg
Updated
Nov 15, 2026, 00:10
VideoLLaMA3-2B
0
Comments
Average rating : 0 (0 users)
Providerdamo-nlp-sg
ModelVideoLLaMA3-2B
Video to Text
wiro playground—damo-nlp-sg/VideoLLaMA3-2B
1 / 1
Maximum 1 video allowed
Video desteklenmiyor

Choose a video that will re-generate

Tell us about any details you want to generate

Your request will cost $0.0006 per second.

Running this model on Wiro costs approximately $0.03033 in total.

(Total cost varies depending on the request's execution time.)
Sample outputs
Loading preview…
damo-nlp-sg-videollama3-2b-sample-5.txt
Loading preview…
damo-nlp-sg-videollama3-2b-sample-6.txt
Updated Nov 15, 2026, 00:10

Overview

This is the 2B video model in DAMO-NLP-SG's VideoLLaMA 3 series. Qwen2.5-1.5B-Instruct handles the language side, and the model card lists VideoLLaMA3-2B-Image as a base model, so the video weights start from the image checkpoint of the same size. The paper describes four training stages: the first three mainly build image understanding and the fourth is dedicated to video.

The model sees a clip as a run of sampled frames. Before those frames reach the language model, a step the authors call the Differential Frame Pruner compares each patch with the same patch in the previous frame and drops it when the pixel difference falls below a threshold. A tripod shot of a quiet scene loses many patches this way, a handheld shot fewer. The authors' stated aim is a compact token sequence that keeps the model on the parts of the video that move.

DAMO-NLP-SG's results table gives this model 59.6 on VideoMME without subtitles and 63.4 with them, 65.4 on MLVU (dev), 57.1 on LongVideoBench (val) and 81.1 on NextQA. MVBench is the only row of that table where a rival 2B model comes out ahead: InternVL2.5 scores 68.8 there against 65.5.

For a single photo, chart or scanned page, use VideoLLaMA3 2B Image, the image checkpoint this model was built from. Every run on this page needs a video, so pick it when the input is a clip.

What you can build

  • A searchable index for a video library, built from a text description of each clip
  • Direct answers about footage, such as who appears in the shot and what they are doing
  • A review pass that flags out-of-place content; on this page it picks out a person vacuuming sand in a desert
  • First-draft summaries of drone shots or wildlife footage for an editor to correct

Inputs

Every run needs a video, either uploaded as a file or passed as a link. The prompt box is prefilled with "Please describe this video in detail." In the advanced settings, max new tokens caps how long the answer can get; it defaults to 128 and accepts values from 10 to 4096.

Outputs

An answer in English text. For an aerial clip of a city skyline, the sample reply is three sentences about a large white-roofed stadium near the center of the frame. The desert clip gets two sentences on why vacuuming sand is unusual.

Recommended settings

The default of 128 tokens covers a short paragraph like the two samples. For a scene-by-scene account of a longer video, raise it to several hundred. Name what you care about in the prompt, such as what the person is holding or what happens after the car stops, since the default prompt asks only for a general description.

Limitations

  • The model works from the frames alone. The card lists no audio input and the paper names audio and speech as unexplored, so speech and music in a clip are ignored.
  • Frames are sampled at intervals, so a very brief event can fall between two of them.
  • The authors point out that video training data has lower annotation quality and less variety than image data, which can hold the model back on unfamiliar kinds of footage.

Safety & compliance

The model card lists the Apache 2.0 license, but the project's GitHub page marks VideoLLaMA 3 as a research preview for non-commercial use only, under the Qwen model licenses, the terms of use of data generated by OpenAI and Gemini, and ShareGPT's privacy practices. Read those terms before any commercial use. Upload footage of identifiable people only with their consent.

Example prompts

Great starting points for VideoLLaMA3-2B.

Describe this video in detail.Video Generation
Please describe the video.Video Generation
Please describe the video and the big structre in it.Video Generation
What is unusual in the video?Video Generation
What is the woman doing in the video?Video Generation

API quick start

Run VideoLLaMA3-2B with a single API call.

POST https://api.wiro.ai/v1/Run/damo-nlp-sg/videollama3-2b
{
  "prompt": "Please describe this video in detail.",
  "inputVideo": "https://your-cdn.com/input.mp4",
  "max_new_tokens": 128
}
curl
curl -X POST "https://api.wiro.ai/v1/Run/damo-nlp-sg/videollama3-2b" \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_WIRO_API_KEY" \
  --data-binary @- <<'JSON'
{
  "prompt": "Please describe this video in detail.",
  "inputVideo": "https://your-cdn.com/input.mp4",
  "max_new_tokens": 128
}
JSON

Other models you might like

Explore similar video generation models.

View all
Video to Text

ask-video

wiro

VideoLLaMA3-7B
Video to Text

VideoLLaMA3-7B

wiro

Video to Text

cogvlm2-llama3-caption

thudm

video-nsfw-detection
Video to Text

video-nsfw-detection

wiro

  • VideoLLaMA3-2B-Imagedamo-nlp-sg

Discover, test, and run AI models, build workflows and agents with one unified API.

All systems operational
WiroModelsAgentsMCP ServerLLM GatewayBrowser ExtensionIntegrationsPricingAboutBlogCareersContactPartnerChangelogStatusFAQ
CategoriesText to ImageImage to ImageImage EditingImage to TextText to VideoImage to VideoVideo to VideoText to SpeechSpeech to TextSpeech to SpeechVoice CloneRealtime ConversationMusic GenerationLLMUtilities
ModelsNano Banana 2GPT Image 2.5Seedream V5 ProSeedance 2.5Veo 3.1Kling V3FLUX 3FLUX.2 ProWan 3.0 PrimeGrok Imagine 1.5
PartnersGoogleOpenAIByteDanceBlack Forest LabsKling AIQwenAlibabaxAIMiniMaxElevenLabs
Getting StartedWiro ModelsWiro EffectsIntroductionAuthenticationProjectsCode ExamplesWiro MCP ServerSelf-Hosted MCPn8n IntegrationLLMs.txt
API ReferenceModelsRun a ModelModel ParametersTasksLLM & Chat StreamingWebSocketRealtime VoiceFiles
© 2026 Wiro AI. All rights reserved.
PrivacyTermsData Deletion