Noometry

Alibaba (Qwen), open weights

Qwen2-72B

Qwen2-72B by Alibaba (Qwen) ranks 300th of 354 ranked models on the Noometry Index as of October 2026, with a score of 30.0. Its strongest category is agentic & tool use, where it ranks 146th.

Last verified

Noometry rank
#300 of 354
Index score
30.0
Evidence
Confirmed 26 results
Provider
Alibaba (Qwen)
Released
June 7, 2024
Weights
Open weights
Reasoning
Unknown
Context window
—
Max output
—
Input price
Not listed
Output price
Not listed
Blended price
Not listed
Output speed
Not measured
Value
Not ranked
Knowledge cutoff
Unknown

Category scores

Each category score combines every public result we have in that category.

  1. Coding 29.1
  2. Agentic & Tool Use 17.0
  3. Reasoning 23.2
  4. Math 30.2
  5. Knowledge 21.2
  6. Multilingual 35.9
  7. Instruction Following 61.7
  8. Long Context 36.1
  9. Writing & Preference 40.8
CategoryScoreRankResults
Coding29.1#3104
Agentic & Tool Use17.0#1461
Reasoning23.2#1811
Math30.2#2362
Knowledge21.2#2752
Multilingual35.9#2441
Instruction Following61.7#2411
Long Context36.1#2351
Writing & Preference40.8#2413

Strengths and weaknesses

Categories where Qwen2-72B places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Reasoning23.2−0.4#181 of 350, top 52%
Math30.2−6.4#236 of 327, top 73%
Writing & Preference40.8−13.0#241 of 312, top 78%

Weakest categories

CategoryScorevs medianRank
Agentic & Tool Use17.0−13.3#146 of 154, top 95%
Coding29.1−9.6#310 of 340, top 92%
Knowledge21.2−16.1#275 of 314, top 88%

Closest competitors

The models ranked just above and below Qwen2-72B. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Magistral Small#29630.2$0.750Compare
Qwen2.5 32B Instruct#29730.1$1.22—Compare
Palm 2#29830.0——Compare
Gemma 7B#29930.0——Compare
Gemini 1.5 Flash 8B#30129.9——Compare
Qwen2.5-VL 72B Instruct#30229.9$4.2043Compare
Mistral#30329.9——Compare
OLMo 2 Furious 13B#30429.7——Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
WeirdML11.3%#112 of 119, top 95%Epoch AI
BigCodeBench Instruct38.5%#37 of 64, top 58%BigCodeBench2024-06-07
LMArena Coding1196#243 of 294, top 83%LMArena2026-10-08
BigCodeBench Complete54%#23 of 66, top 35%BigCodeBench2024-06-07

Agentic & Tool Use

BenchmarkScorePositionSettingSourceDate
TheAgentCompany1.1%#14 of 14, top 100%Epoch AI
METR Time Horizons29.9%#30 of 32, top 94%Epoch AI

Reasoning

BenchmarkScorePositionSettingSourceDate
LMArena Hard Prompts1191#243 of 297, top 82%LMArena2026-10-08
Epoch Capabilities Index125.28#157 of 213, top 74%Epoch AI2024-06-07

Math

BenchmarkScorePositionSettingSourceDate
LMArena Math1235#222 of 285, top 78%LMArena2026-10-08
MATH Level 539.1%#53 of 79, top 68%Epoch AI2025-01-27

Knowledge

BenchmarkScorePositionSettingSourceDate
GPQA Diamond40.8%#148 of 186, top 80%Epoch AI2025-01-27
LMArena Expert1171#231 of 273, top 85%LMArena2026-10-08
MMLU82.4%#11 of 81, top 14%Epoch AI

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1176#244 of 297, top 83%LMArena2026-10-08
LMArena Chinese1240#216 of 285, top 76%LMArena2026-10-08
LMArena French1170#199 of 223, top 90%LMArena2026-10-08
LMArena German1151#202 of 231, top 88%LMArena2026-10-08
LMArena Japanese1111#184 of 211, top 88%LMArena2026-10-08
LMArena Korean1083#190 of 213, top 90%LMArena2026-10-08
LMArena Russian1169#245 of 283, top 87%LMArena2026-10-08
LMArena Spanish1169#201 of 226, top 89%LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1181#242 of 298, top 82%LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1192#244 of 291, top 84%LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1203#243 of 297, top 82%LMArena2026-10-08
LMArena Creative Writing1181#241 of 295, top 82%LMArena2026-10-08
LMArena Multi-Turn1196#238 of 295, top 81%LMArena2026-10-08

Compare Qwen2-72B

Other Alibaba (Qwen) models

Frequently asked questions

How good is Qwen2-72B?

Qwen2-72B by Alibaba (Qwen) ranks 300th of 354 ranked models on the Noometry Index as of October 2026, with a score of 30.0. Its strongest category is agentic & tool use, where it ranks 146th.

Is Qwen2-72B open source?

Yes. Qwen2-72B's weights are downloadable; check the license for commercial terms.

What are Qwen2-72B's strengths and weaknesses?

Relative to other ranked models, Qwen2-72B places best in reasoning, math, writing & preference and lowest in agentic & tool use, coding, knowledge.

What is Qwen2-72B best at?

Its best category is agentic & tool use, where it ranks 146th on Noometry.