Alibaba

Qwen2.5-VL 72B

Vérifié par l’équipe de recherche de GetInfoUsDernière révision : October 2026Politique éditoriale
Vision 72B open sourceOfficial Source Verified ↗
79
Intelligence Index

Le modèle vision-langage open source phare d'Alibaba, leader en compréhension de schémas, OCR de documents et raisonnement visuel complexe.

Technical Specifications

Provider
Alibaba
Category
open-source
Release Date
2024-09
Knowledge Cutoff
June 2024
Architecture
Open-Weights Multimodal Vision-Language Transformer
Parameters
72B
Context Window
131K tokens
Max Output
8K tokens
Cached Input Price
$0 / 1M tokens
Batch API Discount
Standard rate
Consumer Subscription
Free / Self-Hosted
Reasoning Mode
Yes (Chain-of-Thought)
Reasoning Effort Control
Fixed
Tool Calling
Yes
Structured Outputs
Yes
Open Weights
Yes (Open Source / Self-Hostable)
Input Modalities
text, vision
Output Modalities
text
Official Documentation
Visit Official Docs ↗

Data Privacy, Licensing & Deployment Governance

API Data Training Policy
100% Isolated when self-hosted on private GPUs or VPC
Model License & Commercial Rights
Permissive Open-Weights (Apache 2.0 / MIT Commercial)
Deployment & Cloud Availability
Self-Hosted GPU (vLLM / Ollama), Private VPC, or Managed Serverless API
Output Ownership
Full commercial ownership of generated outputs

Official Provider Release Intelligence, API Quotas & Agentic Specs

Compare Multi-Models Side-by-Side →
Official API Model ID (SDK String)
qwen2-5-vl-72b-instruct
Supported SDK & API Protocol
OpenAI-Compatible REST / vLLM / SGLang / HuggingFace TGI
Lifecycle & Deprecation Status
Legacy Stable (Upgrade path available in Alibaba lineup)
Production Rate Limits (RPM & TPM)
Unlimited (Hardware-bound on self-hosted GPU) / 10,000 RPM Cloud · Unlimited local throughput / 10M+ TPM Managed Cloud
Long-Context Pricing Tier (>128K / >200K)
Free across full context window (only KV-cache VRAM scales)
Prompt Cache Write & TTL Policy
Automatic prefix KV-caching in vLLM / SGLang (0 cost)
Computer Use & Browser GUI Agent
Supported (Screen coordinate grounding, browser & desktop GUI action loops)
MCP & Parallel Tool Calling
Native MCP Client/Server ready · Parallel function calling · 100% Strict JSON Schema
Web Search Grounding & Code Sandbox
Supported via external Tavily/Exa Search & E2B/Docker Sandbox tools
Fine-Tuning & Distillation Support
Full Weights SFT, LoRA, QLoRA, DPO & GRPO RL (Unsloth / Axolotl / Llama-Factory)
Multimodal Tokenization Billing
Text BPE (~4 chars/token) · Vision (~258–765 tokens per image tile)
Weights Precision & Quantization
BF16 / FP16, FP8 Dynamic, INT4 AWQ / GPTQ, GGUF (Q4_K_M / Q8_0)
Minimum GPU VRAM Requirement
45 GB VRAM (4-bit Quantized) · 155 GB VRAM (Native BF16)
Supported Inference Engines
vLLM, SGLang, Ollama, llama.cpp, TensorRT-LLM, HuggingFace TGI
System Card & Safety Framework
Open-Weights System Card + Customizable Guardrail Layer (Llama Guard / ShieldGemma)
Enterprise Compliance (SOC2 / HIPAA / ZDR)
100% Air-Gapped On-Premises / Private VPC · Full HIPAA, GDPR & SOC2 Control
Engineering & Pricing Workbench

Qwen2.5-VL 72B Cost, Context, Hardware & API Integration

500 / day
4,000 tok
1,000 tok
Conversation Depth (History Re-billing):1 Turn (Single-Shot)
50%
Estimated Monthly API Cost (30 Days)
Free ($0)
Subscription ($20/mo) vs API

Open-weights or free-tier model: $0 API license cost when self-hosted.

Same Monthly Workload Cost Across Popular Reference Models

Qwen2.5-VL 72B (Active)
$0.00
/ month
Gemini 2.5 Flash
$18.00
/ month
DeepSeek V3
$32.70
/ month
GPT-4o
$300.00
/ month
Claude 3.7 Sonnet
$405.00
/ month

Live Prompt Paste Tokenizer & Cost Counter

Paste real system instructions, JSON schemas, or code to measure BPE token counts across tokenizers.

Chars / Words
175 / 24
OpenAI o200k
~46 tok
Claude BPE
~49 tok
Llama / DeepSeek
~51 tok

Hybrid Prompt Routing Cascade Simulator

Route routine queries to a fast lightweight model ($0.15/1M) and escalate only complex reasoning turns to Qwen2.5-VL 72B:

Cascaded Monthly Bill:$13.50/mo
This model is already priced at or below the fast router baseline.

Benchmark Scores & 6-Axis Capability Radar

Benchmark Encyclopedia
Coding (44)Reasoning (67)Knowledge (78)Speed (37)Context (75)Cost Value (98)
Qwen2.5-VL 72B2026 Frontier Median
SWE-bench Verified
38%
GPQA Diamond
58%
MMLU-Pro
77.5%
HumanEval
86%
Chatbot Arena Elo
1315pts

Strengths

  • haut de gamme-palier à poids ouverts visuel raisonnement et vidéo compréhension
  • Dynamic résolution visuel encoding préservant aspect ratios
  • complet auto-hébergement liberty avec zero propriétaire API données retention

Limitations & Trade-offs

  • Demands multi-GPU matériel (e.g. 2x A100/H100 80GB) pour comfortable inférence
  • inférence débit est slower que hosted léger modèles

Editorial Assessment

Le modèle de vision open source le plus performant rivalisant directement avec les modèles propriétaires.

Best for: OCR de documents et graphiques, raisonnement multimodal et analyse visuelle multilingue

Frequently Asked Questions about Qwen2.5-VL 72B

How much does the Qwen2.5-VL 72B API cost per 1M tokens?+

Qwen2.5-VL 72B by Alibaba is priced at $0 per 1M input tokens and $0 per 1M output tokens. With prompt caching enabled, cached input tokens cost $0 per 1M tokens, and asynchronous Batch API requests receive a 0% discount.

Should I use a $20/month subscription or the Qwen2.5-VL 72B pay-as-you-go API?+

A flat $20/month consumer subscription is predictable for daily interactive chat and manual coding. However, if you are building software, running automated workflows, or sending low-to-moderate volume with prompt caching enabled, the pay-as-you-go API is typically far more cost-effective and offers higher rate limits without manual UI caps.

How many words and pages fit in Qwen2.5-VL 72B's context window?+

Qwen2.5-VL 72B features a 131K tokens context window, which equals approximately 98,304 English words or ~262 standard single-spaced pages in a single prompt. Its maximum output generation per turn is 8,192 tokens.

Can I run Qwen2.5-VL 72B locally on my own GPU (Open Weights)?+

Yes, Qwen2.5-VL 72B (72B) releases open weights that can be self-hosted locally or on private cloud GPUs using vLLM, Ollama, SGLang, or llama.cpp.

Does Alibaba train on data sent to the Qwen2.5-VL 72B API, and is commercial use allowed?+

100% Isolated when self-hosted on private GPUs or VPC. License tier: Permissive Open-Weights (Apache 2.0 / MIT Commercial). Full commercial ownership of generated outputs. Supported deployment channels include: Self-Hosted GPU (vLLM / Ollama), Private VPC, or Managed Serverless API.

How does Qwen2.5-VL 72B perform on coding and reasoning benchmarks?+

Qwen2.5-VL 72B achieves a composite Intelligence Index of 79/100, scoring 38% on SWE-bench Verified, 58% on GPQA Diamond, 77.5% on MMLU-Pro, and 1315 pts on Chatbot Arena Elo.

What is Qwen2.5-VL 72B best used for?+

Qwen2.5-VL 72B is best suited for: OCR de documents et graphiques, raisonnement multimodal et analyse visuelle multilingue. Assessment: Le modèle de vision open source le plus performant rivalisant directement avec les modèles propriétaires.

Les scores de benchmarks, tarifs par token et limites évoluent rapidement. Vérifiez toujours les spécifications actuelles sur la documentation officielle du fournisseur.

Sources: Qwen Team Official GitHub & Hugging Face Release↗