LLM Reference

Qwen2.5-72B-Instruct

Released
2024-06-07
Last refreshed
2026-05-22
Status
Researched 46d ago
CodingRAGLong contextClassificationJSON / Tool use

Qwen2.5-72B-Instruct is worth evaluating for coding, rag, and long context when its provider route and context window match the workload.

Use it for

  • Teams evaluating coding, rag, and long context
  • Workloads that can use a 128k context window
  • Buyers comparing 4 tracked provider routes

Do not use it for

  • Vision or document-understanding workloads
Specifications
Family
Qwen2.5
Released
2024-06-07
Context
128k
Parameters
72.7B
Architecture
Decoder Only
Specialization
general
Training
finetuned
Fine-tuning
instruct
Created by

AI research institute of Alibaba Group.

Hangzhou, Zhejiang, China
Founded 2017
Website
Pricing
Output / 1M
$0.280
Input / 1M
$0.280

Cheapest of 7 routes · SiliconFlow

About

Instruction-optimized flagship variant for demanding production applications requiring high-accuracy complex problem-solving across industries.

Qwen2.5-72B-Instruct is a model in the Qwen2.5 family. The structured metadata tracks a 128k-token context window and structured outputs. This page tracks provider routes through DeepInfra, OpenRouter, Fireworks AI, and 4 more, with the cheapest tracked route listed at $0.18 input and $0.54 output per 1M tokens. Headline tracked benchmarks include Google-Proof Q&A 38.4, HellaSwag 95.6, and HumanEval 86.6.

Top use-case fit: coding, agents, and build tasks

Coding

Q/$ B

1 relevant benchmark in the decision map.

RAG

Included by capability and metadata signals in the decision map.

Long context

Included by capability and metadata signals in the decision map.

Provider price ladder

Compare all 7

Compare API pricing across 4 providers for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MRoute
SiliconFlow$0.280$0.280
Serverless
DeepInfra$0.360$0.400
Serverless
OpenRouter$0.360$0.400
Serverless
Novita AI$0.380$0.400
Serverless

Capabilities

Structured Outputs

Benchmark peer barsfor Coding

Benchmark scores(5)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionSource
Google-Proof Q&A38.4diamondhttps://huggingface.co/datasets/open-llm-leaderboard/Qwen__Qwen2.5-72B-Instruct-details
HellaSwag95.6standardhttps://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
HumanEval86.6pass@1https://qwenlm.github.io/blog/qwen2.5-llm/
Massive Multitask Language Understanding88.25-shothttps://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
Chatbot Arena1270.0https://lmarena.ai

Migration checks

No linked migration route is available for this model yet.

Rankings & picks(10)

Comparison and alternatives

Browse all comparisons →
Show all 23 popular comparisonssorted by 7-day search impressions