Model comparison

phi-3-medium 14B vs Qwen-7B

phi-3-medium 14B has enough public results to be ranked (#306); Qwen-7B does not yet, so treat this comparison as directional.

Last verified . 4 shared benchmarks.

phi-3-medium 14B Microsoft

29.1

Rank #306 Reported

Summary

  • They share 4 benchmarks with published results for both.

Side by side

phi-3-medium 14B and Qwen-7B specifications
phi-3-medium 14BQwen-7B
ProviderMicrosoftAlibaba (Qwen)
Noometry Index29.1—
Released2024-04-232023-09-28
WeightsOpenOpen
Context window——
Max output——
Input $ / M tokens——
Output $ / M tokens——
Results tracked138

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

Category by category

Coding Not comparable

phi-3-medium 14B: 36.3 (#198), Qwen-7B: —

Coding benchmarks
Benchmarkphi-3-medium 14BQwen-7B
BigCodeBench Instruct37.6%—
BigCodeBench Complete48.7%—

Reasoning Not comparable

phi-3-medium 14B: —, Qwen-7B: —

Reasoning benchmarks
Benchmarkphi-3-medium 14BQwen-7B
BIG-Bench Hard81.4%45%
Epoch Capabilities Index121.18106.47
Adversarial NLI55.8%—
HellaSwag82.4%—
LAMBADA—67.9%
PIQA—77.9%
WinoGrande81.5%—

Math Not comparable

phi-3-medium 14B: 25.2 (#249), Qwen-7B: —

Math benchmarks
Benchmarkphi-3-medium 14BQwen-7B
MATH Level 517.6%—
GSM8K—51.7%

Knowledge Not comparable

phi-3-medium 14B: 9.1 (#301), Qwen-7B: —

Knowledge benchmarks
Benchmarkphi-3-medium 14BQwen-7B
ARC (AI2) Challenge91.6%75.3%
MMLU78%45%
GPQA Diamond27.6%—
BoolQ—76.4%
OpenBookQA87.4%—
TriviaQA73.9%—

Frequently asked questions

Is phi-3-medium 14B better than Qwen-7B?

phi-3-medium 14B has enough public results to be ranked (#306); Qwen-7B does not yet, so treat this comparison as directional.

How many benchmarks do phi-3-medium 14B and Qwen-7B share?

4 benchmarks have published results for both models. phi-3-medium 14B has 13 scored results on Noometry and Qwen-7B has 8.

Related comparisons

Go deeper