Qwen2-VL-72B-Instruct

Facts

Released

2024-09-17 the day the repository was first published on Hugging Face Hugging Face, read 2026-09-27

Licence

other the licence the model card declares Hugging Face, read 2026-09-27

Open weights

Yes the weights are published in this Hugging Face repository Hugging Face, read 2026-09-27

Parameters

73.4 billion counted from the safetensors weight files Hugging Face, read 2026-09-27

Active parameters

Unknown

Knowledge cutoff

Unknown

Context window

• 32,768 tokens the maximum position embeddings in the model configuration model configuration, read 2026-09-27

Longest output

Unknown

Tool calling

• No the chat template in the tokenizer configuration has no place for tool definitions Hugging Face, read 2026-09-27

Structured output

Unknown

Reasoning controls

Unknown

Inputs and outputs

text, image in, text out Hugging Face, read 2026-09-27

Good for

• Vision: Hugging Face files it under image-text-to-text Hugging Face, read 2026-09-27

Prices

Run it on your own hardware

Weights are the sizes of the files a source lists, or an estimate from the parameter count where none does. Memory is an estimate: weights plus KV cache plus 512 MiB and 5 percent of the weights for runtime buffers. Check it against your hardware.

Start it with a local runtime

• Ollama

  ollama run hf.co/bartowski/Qwen2-VL-72B-Instruct-GGUF:Q4_0

• LM Studio

  lms get bartowski/Qwen2-VL-72B-Instruct-GGUF
  lms server start

• llama.cpp server

  llama-server -hf bartowski/Qwen2-VL-72B-Instruct-GGUF:Q4_0 --jinja

• vLLM

  vllm serve Qwen/Qwen2-VL-72B-Instruct

• SGLang

  sglang serve --model-path Qwen/Qwen2-VL-72B-Instruct --port 30000

Use it from your harness

Set up for Nebius Token Factory with the model Qwen/Qwen2-VL-72B-Instruct. Each endpoint page has the same setup for its own address.

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "nebius": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Nebius Token Factory",
      "options": {
        "baseURL": "https://api.tokenfactory.nebius.com/v1",
        "apiKey": "{env:NEBIUS_API_KEY}"
      },
      "models": {
        "Qwen/Qwen2-VL-72B-Instruct": {
          "name": "Qwen2-VL-72B-Instruct"
        }
      }
    }
  }
}

{
  "providers": {
    "nebius": {
      "baseUrl": "https://api.tokenfactory.nebius.com/v1",
      "api": "openai-completions",
      "apiKey": "$NEBIUS_API_KEY",
      "models": [
        {
          "id": "Qwen/Qwen2-VL-72B-Instruct"
        }
      ]
    }
  }
}

Other ways to reach it

• Together AI, model Qwen/Qwen2-VL-72B-Instruct
• Ollama, model hf.co/bartowski/Qwen2-VL-72B-Instruct-GGUF:Q4_0

Published results

These are results other people published. Baltor did not run them and does not rank models by them.

• Results the maker published on the model card published by Qwen Hugging Face, read 2026-09-27

Two more publishers are linked, never copied, because their terms do not allow it:

• Artificial Analysis publishes its own intelligence, speed and price measurements.
• OpenRouter lists current prices and uptime for each provider it routes to.