all the models — AI benchmark observatory
← Models

Phi-3.5-mini-instruct

Microsoft · open weight · phi-3.5-mini-instruct

Phi-3.5-mini-instruct is a 3.8B-parameter model that supports up to 128K context tokens, with improved multilingual capabilities across over 20 languages. It underwent additional training and safety post-training to enhance instruction-following, reasoning, math, and code generation. Ideal for environments with memory or latency constraints, it uses an MIT license.

RepoQAMMLUHumanEvalMATHMMLU-ProArena Hard

Benchmark scores

BenchmarkScore
RepoQA0.77
MMLU0.69
HumanEval0.63
MATH0.48
MMLU-Pro0.47
Arena Hard0.37
GPQA0.30

Pricing

  • No provider pricing.

AA metrics

No Artificial Analysis link yet.

Arena Elo

  • No Arena snapshot linked.