llmboard.ai
Benchmarks
CompareRankings
llmboard.ai
Benchmarks
CompareRankings
HomeModelsQwen2.5 72B

Alibaba Cloud / Qwen Team model product

Qwen2.5 72B

Qwen2.5-72B-Instruct is an instruction-tuned 72 billion parameter language model, part of the Qwen2.5 series. It is designed to follow instructions, generate long texts (over 8K tokens), understand structured data (e.g., tables), and generate structured outputs, especially JSON. The model supports multilingual capabilities across over 29 languages.

Updated Aug 10, 2026. Default version: Qwen2.5 72B Instruct

Compare
LLMBoard score53.1Qwen2.5 72B Instruct
Coverage80%10 benchmark families
Context window131.1KTokens
Official input price$1.4Alibaba API

On this page

  • Specification
  • Capability
  • Benchmarks
  • Arena
  • Pricing
  • Versions
  • About
  • Compare
  • Similar models
  • FAQ

Model specification

Structured fields from the published default version.

Version
Qwen2.5 72B Instruct
Released
Sep 19, 2024
Knowledge cutoff
Unknown
Parameters
72.7B
Context window
131.1K
Max output
8.2K
Inputs
text
Outputs
text
Open weights
No
License
Qwen

Capability profile

This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.

Qwen2.5 72B Instruct category scores

Benchmark results

Published benchmark records for the scored version Qwen2.5 72B Instruct.

14 rows
Columns

Show columns

AlignBench0.814100.0%CAug 7, 2026
MT-Bench0.9112100.0%CAug 7, 2026
MBPP0.953788.9%CAug 7, 2026
Arena Hard0.862680.0%CAug 7, 2026
MultiPL-E0.871350.0%CAug 7, 2026
GSM8k1.0104880.8%CAug 7, 2026
MATH0.8167178.6%CAug 7, 2026
MMLU-Redux0.9324834.0%CAug 7, 2026
HumanEval0.9337657.3%CAug 7, 2026
LiveBench0.535388.1%CAug 7, 2026
LiveCodeBench0.6377350.0%CAug 7, 2026
IFEval0.8426535.9%CAug 7, 2026
MMLU-Pro0.78112937.5%CAug 7, 2026
GPQA0.518123322.4%CAug 7, 2026

Arena results

Preference and agent-evaluation signals from published Arena datasets.

56 rows
Columns

Show columns

textjapanese1901180.2788N/AAug 6, 2026
textfrench1971278.8340N/AAug 6, 2026
text style controljapanese2001210.6788N/AAug 6, 2026
textkorean2041188.1527N/AAug 6, 2026
text style controlfrench2061319.7340N/AAug 6, 2026
textspanish2091254.3347N/AAug 6, 2026
text style controlkorean2091229.3527N/AAug 6, 2026
textindustry mathematical2151290.54,565N/AAug 6, 2026
text style controlspanish2151291.2347N/AAug 6, 2026
textgerman2181233.4926N/AAug 6, 2026
textmath2221282.65,415N/AAug 6, 2026
text style controlgerman2221265.6926N/AAug 6, 2026
text style controlindustry mathematical2221309.64,565N/AAug 6, 2026
textchinese2241271.53,329N/AAug 6, 2026
text style controlchinese2261321.23,329N/AAug 6, 2026
textexpert2271244.52,397N/AAug 6, 2026
text style controlindustry legal and government2271330.52,386N/AAug 6, 2026
text style controlmath2271296.35,415N/AAug 6, 2026
textindustry legal and government2281288.02,386N/AAug 6, 2026
textcoding2311292.96,688N/AAug 6, 2026
textlonger query2331282.06,008N/AAug 6, 2026
textrussian2331263.15,776N/AAug 6, 2026
textindustry software and it services2341293.210,448N/AAug 6, 2026
text style controlcoding2341355.56,688N/AAug 6, 2026
texthard prompts2351271.010,545N/AAug 6, 2026
text style controlrussian2371296.45,776N/AAug 6, 2026
textindustry medicine and healthcare2381257.71,956N/AAug 6, 2026
text style controlexpert2381293.82,397N/AAug 6, 2026
textinstruction following2391254.416,363N/AAug 6, 2026
textindustry business and management and financial operations2401252.64,656N/AAug 6, 2026
text style controlindustry software and it services2401343.410,448N/AAug 6, 2026
textmulti turn2411271.87,250N/AAug 6, 2026
text style controllonger query2411317.06,008N/AAug 6, 2026
text style controlmulti turn2411298.87,250N/AAug 6, 2026
texthard prompts english2421281.36,208N/AAug 6, 2026
textindustry life and physical and social science2441275.36,932N/AAug 6, 2026
textnon english2451251.618,845N/AAug 6, 2026
text style controlhard prompts english2451327.86,208N/AAug 6, 2026
text style controlinstruction following2451292.216,363N/AAug 6, 2026
text style controlhard prompts2461317.510,545N/AAug 6, 2026
text style controlindustry business and management and financial operations2461299.24,656N/AAug 6, 2026
text style controlindustry medicine and healthcare2461312.81,956N/AAug 6, 2026
text style controlnon english2471284.918,845N/AAug 6, 2026
textexclude ties2501189.724,444N/AAug 6, 2026
textoverall2511269.239,406N/AAug 6, 2026
textenglish2531282.820,561N/AAug 6, 2026
text style controlindustry life and physical and social science2531318.36,932N/AAug 6, 2026
text style controlindustry writing and literature and language2531282.010,827N/AAug 6, 2026
text style controlexclude ties2551237.724,444N/AAug 6, 2026
text style controloverall2571302.939,406N/AAug 6, 2026
textindustry writing and literature and language2581246.910,827N/AAug 6, 2026
text style controlenglish2601316.220,561N/AAug 6, 2026
text style controlcreative writing2631254.35,684N/AAug 6, 2026
textcreative writing2641221.65,684N/AAug 6, 2026
textindustry entertainment and sports and media2661211.66,207N/AAug 6, 2026
text style controlindustry entertainment and sports and media2741244.96,207N/AAug 6, 2026

Pricing

Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.

Official API
$1.4 input, $5.6 output per 1M
Official provider
Alibaba
Lowest third-party
From $0.11 input, $0.38 output per 1M via Abacus
Tracked offerings
6
6 rows
Columns

Show columns

AbacusQwen/Qwen2.5-72B-Instructglobal$0.11$0.38128KAug 7, 2026
Kilo Gatewayqwen/qwen-2.5-72b-instructglobal$0.36$0.4032.8KAug 7, 2026
OpenRouterqwen/qwen-2.5-72b-instructglobal$0.36$0.4032.8KAug 7, 2026
NovitaAIqwen/qwen-2.5-72b-instructglobal$0.38$0.4032KAug 7, 2026
Alibaba (China)qwen2-5-72b-instructglobal$0.574$1.72131.1KAug 7, 2026
Alibabaqwen2-5-72b-instructglobal$1.4$5.6131.1KAug 7, 2026

Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.

Qwen2.5 72B versions

All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.

1 rows
Columns

Show columns

Qwen2.5 72B InstructSep 19, 202453.172.7B131.1K8.2KNoQwen

What is Qwen2.5 72B?

A concise description based on the published model registry.

Qwen2.5-72B-Instruct is an instruction-tuned 72 billion parameter language model, part of the Qwen2.5 series. It is designed to follow instructions, generate long texts (over 8K tokens), understand structured data (e.g., tables), and generate structured outputs, especially JSON. The model supports multilingual capabilities across over 29 languages.

Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.

Data snapshot: 2026-08-07. Editorial model content is not available in the backend.

Qwen2.5 72B vs nearby models

Open a comparison with the three ranked models immediately above and below this model.

Qwen2.5 72BvsLlama 3.3 70BQwen2.5 72BvsQwen3 Next 80B A3B ThinkingQwen2.5 72BvsQwen3.5 35B A3BQwen2.5 72BvsGemini 2.0 FlashQwen2.5 72BvsSarvam 105BQwen2.5 72BvsLongCat Flash Chat

Models similar to Qwen2.5 72B

Recommendations prioritize the same model type and family, then the closest published LLMBoard score.

#71+0.2
AC

Qwen3.5 35B A3B

Alibaba Cloud / Qwen Team

53.3 LLMBoard

DetailsCompare
#70+1.1
AC

Qwen3 Next 80B A3B Thinking

Alibaba Cloud / Qwen Team

54.1 LLMBoard

DetailsCompare
#78-1.3
AC

Qwen3 VL 32B Thinking

Alibaba Cloud / Qwen Team

51.7 LLMBoard

DetailsCompare
#79-1.5
AC

Qwen2.5 VL 32B

Alibaba Cloud / Qwen Team

51.5 LLMBoard

DetailsCompare
#59+3.9
AC

Qwen2.5 32B

Alibaba Cloud / Qwen Team

56.9 LLMBoard

DetailsCompare
#86-4.0
AC

Qwen3 235B A22B

Alibaba Cloud / Qwen Team

49.0 LLMBoard

DetailsCompare

FAQ

Common questions about Qwen2.5 72B.

When was Qwen2.5 72B released?

Qwen2.5 72B's default version was released on Sep 19, 2024.

How much does Qwen2.5 72B cost?

Qwen2.5 72B's official API price is $1.4 per million input tokens and $5.6 per million output tokens via Alibaba. The lowest tracked third-party offer starts at $0.11 input and $0.38 output via Abacus.

Who created Qwen2.5 72B?

Qwen2.5 72B is published under Alibaba Cloud / Qwen Team in the model registry.

What is the context window for Qwen2.5 72B?

The default version has a 131.1K token context window.

Is Qwen2.5 72B open weight?

No. The default version is not marked as having publicly available weights.

How many API providers offer Qwen2.5 72B?

6 published provider offerings are linked to the default version.

What models should I compare Qwen2.5 72B with?

Nearby ranked alternatives include Llama 3.3 70B, Qwen3 Next 80B A3B Thinking, Qwen3.5 35B A3B.

Rankings

OverallCodingText ArenaPricing

Modalities

Image GenerationVideo GenerationSpeech-to-TextEmbeddings

Benchmarks

All BenchmarksReasoningMathCoding

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai