Noometry

Model comparison

Mixtral 8x22B vs Qwen2.5 72B Instruct

Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 27.1 on the Noometry Index.

Last verified . 31 shared benchmarks.

Mixtral 8x22B Mistral AI

27.1

Rank #333 Confirmed

Qwen2.5 72B Instruct Alibaba (Qwen)

31.9

Rank #267 Confirmed

Summary

  • They share 31 benchmarks with published results for both. Mixtral 8x22B scores higher in 2 categories and Qwen2.5 72B Instruct in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Qwen2.5 72B Instruct leads 27.0 to 15.1.
  • The biggest single-benchmark swing is MATH Level 5: 24.2% for Mixtral 8x22B and 63.2% for Qwen2.5 72B Instruct.
  • Qwen2.5 72B Instruct is cheaper at $1.40 / $5.60 per million input/output tokens, against $2 / $6 for Mixtral 8x22B.
  • Qwen2.5 72B Instruct accepts more context: 131K tokens versus 64K.

Side by side

Mixtral 8x22BQwen2.5 72B Instruct
ProviderMistral AIAlibaba (Qwen)
Noometry Index27.131.9
Released2024-04-172024-09
WeightsOpenOpen
Context window64K131K
Max output64K8K
Input $ / M tokens$2$1.40
Output $ / M tokens$6$5.60
Results tracked3443

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

Category by category

Coding Qwen2.5 72B Instruct leads

Mixtral 8x22B: 24.2 (#329), Qwen2.5 72B Instruct: 33.2 (#260)

BenchmarkMixtral 8x22BQwen2.5 72B Instruct
WeirdML3.2%16%
BigCodeBench Instruct40.6%45.8%
LMArena Coding11661292
BigCodeBench Complete50.2%55.9%
HumanEval+72%—
MBPP+64.3%—

Agentic & Tool Use Too close to call

Mixtral 8x22B: 23.1 (#127), Qwen2.5 72B Instruct: 22.1 (#133)

BenchmarkMixtral 8x22BQwen2.5 72B Instruct
TheAgentCompany—5.7%
Cybench7.5%—
BALROG—16.2%
METR Time Horizons—35.8%

Reasoning Qwen2.5 72B Instruct leads

Mixtral 8x22B: 19.9 (#248), Qwen2.5 72B Instruct: 22.3 (#199)

BenchmarkMixtral 8x22BQwen2.5 72B Instruct
LMArena Hard Prompts11501271
DTBench55.1%62.9%
Epoch Capabilities Index122.03129
ForecastBench56.357.5
LMCA—13.4%
BIG-Bench Hard—79.8%
HellaSwag—84.8%
PIQA—82.6%
WinoGrande—82.3%

Math Mixtral 8x22B leads

Mixtral 8x22B: 22.9 (#275), Qwen2.5 72B Instruct: 19.3 (#287)

BenchmarkMixtral 8x22BQwen2.5 72B Instruct
Omni-MATH16.3%33%
LMArena Math11841283
MATH Level 524.2%63.2%
OTIS Mock AIME 2024-2025—8.1%

Knowledge Qwen2.5 72B Instruct leads

Mixtral 8x22B: 15.1 (#293), Qwen2.5 72B Instruct: 27.0 (#253)

BenchmarkMixtral 8x22BQwen2.5 72B Instruct
GPQA Diamond34.1%49.1%
MMLU-Pro46%63.1%
GPQA (HELM)33.4%42.6%
LMArena Expert11131245
MMLU77.8%85.3%
Confabulations—19.1%
ARC (AI2) Challenge—94.5%
TriviaQA—71.9%

Multilingual Qwen2.5 72B Instruct leads

Mixtral 8x22B: 32.8 (#255), Qwen2.5 72B Instruct: 41.0 (#213)

BenchmarkMixtral 8x22BQwen2.5 72B Instruct
LMArena Non-English11281252
LMArena Chinese11161272
LMArena French11661280
LMArena German11411234
LMArena Japanese10371180
LMArena Korean10571188
LMArena Russian11581264
LMArena Spanish11511256

Instruction Following Qwen2.5 72B Instruct leads

Mixtral 8x22B: 57.7 (#266), Qwen2.5 72B Instruct: 65.5 (#221)

BenchmarkMixtral 8x22BQwen2.5 72B Instruct
IFEval72.4%80.6%
LMArena Instruction Following11471254

Long Context Qwen2.5 72B Instruct leads

Mixtral 8x22B: 34.7 (#247), Qwen2.5 72B Instruct: 38.9 (#188)

BenchmarkMixtral 8x22BQwen2.5 72B Instruct
LMArena Longer Query11441282

Writing & Preference Qwen2.5 72B Instruct leads

Mixtral 8x22B: 36.9 (#262), Qwen2.5 72B Instruct: 46.7 (#215)

BenchmarkMixtral 8x22BQwen2.5 72B Instruct
LMArena Text11621269
LMArena Creative Writing11411221
WildBench71.1%80.2%
LMArena Multi-Turn11301272

Frequently asked questions

Is Mixtral 8x22B better than Qwen2.5 72B Instruct?

Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 27.1 on the Noometry Index.

Which is cheaper, Mixtral 8x22B or Qwen2.5 72B Instruct?

Qwen2.5 72B Instruct is cheaper. It lists at $1.40 per million input tokens and $5.60 per million output tokens; Mixtral 8x22B lists at $2 and $6.

Is Mixtral 8x22B or Qwen2.5 72B Instruct better for coding?

Qwen2.5 72B Instruct scores higher on coding benchmarks: 33.2 versus 24.2 in the Noometry coding category.

Which has the bigger context window?

Qwen2.5 72B Instruct does, with 131K tokens against 64K.

How many benchmarks do Mixtral 8x22B and Qwen2.5 72B Instruct share?

31 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Qwen2.5 72B Instruct has 43.

Related comparisons

Go deeper