Noometry

Model comparison

Mistral Small vs Qwen2-72B

Mistral Small is the stronger model overall, scoring 33.4 to 30.0 on the Noometry Index.

Last verified . 22 shared benchmarks.

Mistral Small Mistral AI

33.4

Rank #243 Confirmed

Qwen2-72B Alibaba (Qwen)

30.0

Rank #300 Confirmed

Summary

  • They share 22 benchmarks with published results for both. Mistral Small scores higher in 7 categories and Qwen2-72B in 2 categories; 9 gaps are clear of the uncertainty.
  • The widest gap is in math, where Qwen2-72B leads 30.2 to 16.4.
  • The biggest single-benchmark swing is MATH Level 5: 46.8% for Mistral Small and 39.1% for Qwen2-72B.

Side by side

Mistral SmallQwen2-72B
ProviderMistral AIAlibaba (Qwen)
Noometry Index33.430.0
Released2024-02-262024-06-07
WeightsOpenOpen
Context window262K—
Max output256K—
Input $ / M tokens$0.15—
Output $ / M tokens$0.60—
Results tracked3926

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

Category by category

Coding Mistral Small leads

Mistral Small: 34.0 (#247), Qwen2-72B: 29.1 (#310)

BenchmarkMistral SmallQwen2-72B
BigCodeBench Instruct36.1%38.5%
LMArena Coding13621196
BigCodeBench Complete46.6%54%
SciCode26.5%—
WeirdML—11.3%
LiveBench Coding36.2%—
ALE-Bench497.62—

Agentic & Tool Use Mistral Small leads

Mistral Small: 28.1 (#93), Qwen2-72B: 17.0 (#146)

BenchmarkMistral SmallQwen2-72B
Berkeley Function Calling Leaderboard37.1%—
TheAgentCompany—1.1%
METR Time Horizons—29.9%

Reasoning Qwen2-72B leads

Mistral Small: 19.8 (#250), Qwen2-72B: 23.2 (#181)

BenchmarkMistral SmallQwen2-72B
LMArena Hard Prompts13351191
Kagi LLM Benchmark37.8%—
CritPt0%—
LiveBench Reasoning44.8%—
DTBench70.9%—
LiveBench Data Analysis53.7%—
LMCA20.6%—
Epoch Capabilities Index—125.28
LiveBench44%—

Math Qwen2-72B leads

Mistral Small: 16.4 (#293), Qwen2-72B: 30.2 (#236)

BenchmarkMistral SmallQwen2-72B
LMArena Math13411235
MATH Level 546.8%39.1%
OTIS Mock AIME 2024-20255.8%—
LiveBench Math39.9%—

Knowledge Mistral Small leads

Mistral Small: 31.0 (#222), Qwen2-72B: 21.2 (#275)

BenchmarkMistral SmallQwen2-72B
GPQA Diamond47.5%40.8%
LMArena Expert12911171
MMLU68.7%82.4%
Vectara Hallucination Rate5.1%—

Multimodal Not comparable

Mistral Small: 33.5 (#96), Qwen2-72B: —

BenchmarkMistral SmallQwen2-72B
LMArena Vision1142—

Multilingual Mistral Small leads

Mistral Small: 45.5 (#169), Qwen2-72B: 35.9 (#244)

BenchmarkMistral SmallQwen2-72B
LMArena Non-English13151176
LMArena Chinese13401240
LMArena French13371170
LMArena German13401151
LMArena Japanese12751111
LMArena Korean12591083
LMArena Russian13241169
LMArena Spanish13461169

Instruction Following Mistral Small leads

Mistral Small: 66.4 (#209), Qwen2-72B: 61.7 (#241)

BenchmarkMistral SmallQwen2-72B
LMArena Instruction Following13101181
LiveBench Instruction Following63.7%—

Long Context Mistral Small leads

Mistral Small: 40.4 (#156), Qwen2-72B: 36.1 (#235)

BenchmarkMistral SmallQwen2-72B
LMArena Longer Query13271192

Writing & Preference Mistral Small leads

Mistral Small: 52.5 (#171), Qwen2-72B: 40.8 (#241)

BenchmarkMistral SmallQwen2-72B
LMArena Text13381203
LMArena Creative Writing13051181
LMArena Multi-Turn13441196
LiveBench Language30.5%—

Frequently asked questions

Is Mistral Small better than Qwen2-72B?

Mistral Small is the stronger model overall, scoring 33.4 to 30.0 on the Noometry Index.

Is Mistral Small or Qwen2-72B better for coding?

Mistral Small scores higher on coding benchmarks: 34.0 versus 29.1 in the Noometry coding category.

How many benchmarks do Mistral Small and Qwen2-72B share?

22 benchmarks have published results for both models. Mistral Small has 39 scored results on Noometry and Qwen2-72B has 26.

Related comparisons

Go deeper