Language modelActive

Qwen 2.5 72B

Alibaba

Released: September 2024
Data date: September 8, 2026
Successor: Qwen 3 32B

Price: $0.36 · $0.4
Input · Output / 1M tokens
Input: Text
Output: Text

New in Qwen 2.5 72B over Qwen2 is the combination of broader knowledge, stronger instruction following, longer text generation, and structured JSON output. The 72B variant remains a dense, text-only instruct model without a Mixture-of-Experts design.

The open repository handles up to 131,072 tokens and can generate up to 8,192 tokens. Qwen’s official Qwen2.5 announcement gives up to 18 trillion training tokens for the series, not as a separate figure for this checkpoint. Qwen 3 32B belongs to the later Qwen3 generation and has 32.8 billion parameters according to its model card.

Specifications and access

Specification	Value and source
Weights	Open weightsSource
Access	API, LocalSource
Parameters	72.7BSource
Context window	131,072 tokensSource
Knowledge cutoff	December 2023Source
Weights repository	Qwen/Qwen2.5-72B-InstructSource
License	qwen (The model card declares a non-SPDX custom/other license; inspect the linked LICENSE before publishing a generic license badge.)Source
Architecture	dense transformerSource
Input	TextSource
Output	TextSource
Reasoning mode	instructSource
Tool calling	YesSource
Structured output	YesSource
Native context window	131,072 tokensSource
Maximum context window	131,072 tokensSource
Parameters	72.7 BSource
Training tokens	18 trillionSource
Repository created	2024-09-16 (Repository creation date is a precise publication proxy, but may differ from the public announcement date or an individual checkpoint's first availability.)Source
Maximum output	8,192 tokensSource

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

MMLU (Show measurement, test conditions, and source)

Source value: 86.1
Score: 86.1
Metric: percent
Unit: %
Category: knowledge
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen2.5 Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report/card benchmark table; exact model column

Source Qwen2.5 Technical Report

MMLU-Pro (Show measurement, test conditions, and source)

Source value: 58.1
Score: 58.1
Metric: percent
Unit: %
Category: knowledge
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen2.5 Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report/card benchmark table; exact model column

Source Qwen2.5 Technical Report

HumanEval (Show measurement, test conditions, and source)

Source value: 59.1
Score: 59.1
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen2.5 Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report/card benchmark table; exact model column

Source Qwen2.5 Technical Report

MBPP (Show measurement, test conditions, and source)

Source value: 84.7
Score: 84.7
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen2.5 Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report/card benchmark table; exact model column

Source Qwen2.5 Technical Report

Artificial Analysis Intelligence Index (Show measurement, test conditions, and source)

Source value: 7.73490670575182
Score: 7.73490670575182
Metric: points
Unit: points
Direction: Higher is better
Source type: independent-evaluator
Evaluator: Artificial Analysis
Status: active
Retrieved at: 2026-09-08
Methodology: Independent evaluation. unknown

Source Artificial Analysis

AA-Omniscience Index (Show measurement, test conditions, and source)

Source value: -53.0666666666667
Score: -53.0666666666667
Metric: points
Unit: points
Direction: Higher is better
Source type: independent-evaluator
Evaluator: Artificial Analysis
Status: active
Retrieved at: 2026-09-08
Methodology: Independent evaluation. unknown

Source Artificial Analysis

Predecessor and successor

Sources and data date

Type	Evidence and data date
Price	Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. Source: Pricing · API pricing
Input	Documented modalities. Retrieved September 8, 2026. Source: Model documentation · Model metadata (retrieved September 8, 2026) · Qwen2.5-72B-Instruct model card · Editorial description reviewed October 2, 2026
Output	Documented modalities. Retrieved September 8, 2026. Source: Model documentation · Model metadata (retrieved September 8, 2026) · Qwen2.5-72B-Instruct model card · Editorial description reviewed October 2, 2026
Research status	Research date September 8, 2026. 4 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Not found in the inspected sources	active-parameters-billions Original wording Dense checkpoint; no separate activated-parameter count is applicable/published.
Additional source	Context window data
Additional source	Qwen2.5 Technical Report (retrieved September 8, 2026)
Additional source	Artificial Analysis (retrieved September 8, 2026)
Additional source	Model metadata (retrieved September 8, 2026)
Additional source	Qwen2.5 official release · Editorial description reviewed October 2, 2026
Additional source	Qwen3-32B model card · Editorial description reviewed October 2, 2026
Additional source	OpenRouter: Popularity index (derived), CC BY 4.0
Additional source	Vercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional source	Creative Commons Attribution 4.0
Additional source	Qwen: "Qwen3: Think Deeper, Act Faster" (2026-09-19)

Qwen 2.5 72B

Specifications and access

Measurements without matching peer values

Our tests

Pelican on a bicycle

Popularity

Published prices

Predecessor and successor

Sources and data date