Noometry

Model comparison

Claude 3 Haiku vs Mixtral 8x7B

Mixtral 8x7B is the stronger model overall, scoring 27.1 to 25.9 on the Noometry Index.

Last verified . 26 shared benchmarks.

Claude 3 Haiku Anthropic

25.9

Rank #340 Confirmed

Mixtral 8x7B Mistral AI

27.1

Rank #334 Confirmed

Summary

  • They share 26 benchmarks with published results for both. Claude 3 Haiku scores higher in 4 categories and Mixtral 8x7B in 4 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in instruction following, where Claude 3 Haiku leads 61.3 to 51.0.
  • The biggest single-benchmark swing is GPQA Diamond: 36.3% for Claude 3 Haiku and 30.6% for Mixtral 8x7B.
  • Mixtral 8x7B has downloadable open weights; the other is API-only.

Side by side

Claude 3 HaikuMixtral 8x7B
ProviderAnthropicMistral AI
Noometry Index25.927.1
Released2024-03-072023-12-11
WeightsProprietaryOpen
Context window—32K
Max output—32K
Input $ / M tokens—$0.70
Output $ / M tokens—$0.70
Results tracked3738

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

Category by category

Coding Mixtral 8x7B leads

Claude 3 Haiku: 26.4 (#325), Mixtral 8x7B: 32.8 (#269)

BenchmarkClaude 3 HaikuMixtral 8x7B
LMArena Coding11991126
HumanEval+68.9%39.6%
MBPP+68.8%49.7%
WeirdML9.8%—
BigCodeBench Instruct39.4%—
BigCodeBench Complete50.1%—
CadEval12%—

Reasoning Mixtral 8x7B leads

Claude 3 Haiku: 16.3 (#307), Mixtral 8x7B: 18.2 (#285)

BenchmarkClaude 3 HaikuMixtral 8x7B
LMArena Hard Prompts11741115
DTBench50.1%49.6%
Epoch Capabilities Index118.35118.47
ForecastBench53.256.3
WinoGrande74.2%77.2%
Kagi LLM Benchmark34.2%—
LMCA8.8%—
Adversarial NLI—55.2%
HellaSwag—86.7%
PIQA—83.6%

Math Mixtral 8x7B leads

Claude 3 Haiku: 9.8 (#319), Mixtral 8x7B: 18.8 (#289)

BenchmarkClaude 3 HaikuMixtral 8x7B
LMArena Math11881147
MATH Level 514.9%10%
OTIS Mock AIME 2024-20251.8%—
Omni-MATH—10.5%
GSM8K—74.4%

Knowledge Claude 3 Haiku leads

Claude 3 Haiku: 17.3 (#285), Mixtral 8x7B: 11.0 (#301)

BenchmarkClaude 3 HaikuMixtral 8x7B
GPQA Diamond36.3%30.6%
LMArena Expert11481088
MMLU73.8%70.6%
MMLU-Pro—33.5%
Confabulations34.2%—
GPQA (HELM)—29.6%
ARC (AI2) Challenge—87.3%
OpenBookQA—85.8%
TriviaQA—82.2%

Multimodal Not comparable

Claude 3 Haiku: 23.6 (#128), Mixtral 8x7B: —

BenchmarkClaude 3 HaikuMixtral 8x7B
LMArena Vision950—
ScienceQA72%—

Multilingual Claude 3 Haiku leads

Claude 3 Haiku: 36.0 (#243), Mixtral 8x7B: 29.6 (#266)

BenchmarkClaude 3 HaikuMixtral 8x7B
LMArena Non-English11781077
LMArena Chinese11551055
LMArena French11951166
LMArena German11741114
LMArena Japanese1102931
LMArena Korean1109968
LMArena Russian12041090
LMArena Spanish11661111

Instruction Following Claude 3 Haiku leads

Claude 3 Haiku: 61.3 (#247), Mixtral 8x7B: 51.0 (#297)

BenchmarkClaude 3 HaikuMixtral 8x7B
LMArena Instruction Following11731109
IFEval—57.5%

Long Context Claude 3 Haiku leads

Claude 3 Haiku: 36.1 (#237), Mixtral 8x7B: 33.4 (#260)

BenchmarkClaude 3 HaikuMixtral 8x7B
LMArena Longer Query11901103

Writing & Preference Mixtral 8x7B leads

Claude 3 Haiku: 29.7 (#291), Mixtral 8x7B: 34.2 (#270)

BenchmarkClaude 3 HaikuMixtral 8x7B
LMArena Text11951132
LMArena Creative Writing11571109
LMArena Multi-Turn11901115
EQ-Bench Creative Writing717—
WildBench—67.3%

Frequently asked questions

Is Claude 3 Haiku better than Mixtral 8x7B?

Mixtral 8x7B is the stronger model overall, scoring 27.1 to 25.9 on the Noometry Index.

Is Claude 3 Haiku or Mixtral 8x7B better for coding?

Mixtral 8x7B scores higher on coding benchmarks: 32.8 versus 26.4 in the Noometry coding category.

How many benchmarks do Claude 3 Haiku and Mixtral 8x7B share?

26 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Mixtral 8x7B has 38.

Related comparisons

Go deeper