Qwen2.5-VL 72B — Alibaba | Parasail

Model details

Specs & substance

Spec sheet

• Developed by: Alibaba
• Model family: Qwen
• Category: vision
• Modality: Text + Vision
• Context window: 128K tokens
• Architecture: Dense VL
• Version: VL 72B
• License: Apache 2.0
• Pricing: $0.80 in $1.00 out · $0.40 cache
• Released: Jan 2025
• Endpoint: parasail-qwen25-vl-72b-instruct

Proven vision-language model for OCR, document QA, and visual grounding.

Qwen2.5-VL 72B is part of the Qwen family by Alibaba, categorized as a vision model. It supports a 128K tokens context window and is built on a Dense VL architecture.

Key strengths: vision, document. Parasail serves Qwen2.5-VL 72B on a global fleet of current-gen GPUs behind a single OpenAI-compatible endpoint — with per-token pricing, no minimums, and dedicated capacity options when you need guaranteed throughput.

Integrate

Drop-in via the OpenAI SDK

Point any OpenAI-compatible chat client at Parasail and change the model name. That's it.

Python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.parasail.io/v1"
)

response = client.chat.completions.create(
    model="parasail-qwen25-vl-72b-instruct",
    messages=[
        {"role": "user", "content": "Hello, what can you do?"}
    ],
    stream=True,
    stream_options={"include_usage": True},
    top_p=1,
    max_tokens=1000,
    temperature=1
)

for chunk in response:
    if chunk.choices and chunk.choices[0].delta.content is not None:
        print(chunk.choices[0].delta.content, end="", flush=True)

Bash

curl https://api.parasail.io/v1/chat/completions \
  -H "Authorization: Bearer $PARASAIL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "parasail-qwen25-vl-72b-instruct",
    "messages": [{"role": "user", "content": "Hello, what can you do?"}],
    "stream": true,
    "max_tokens": 1000
  }'