Phi-3-medium-4k-instruct: benchmark scores, price and ranking
Phi-3-medium-4k-instruct from microsoft ranks #321 of 743 LLMs on the LLMs Tiger aggregate benchmark score (48.8), based on 2 benchmark results captured up to 2026-10-13.
- Organisation
- microsoft
- Type
- Open-weight / local model
- Open weights
- Yes
- Cheapest API price
- $0.170 in / $0.680 out per million tokens (via azure_ai, observed 2026-10-13T20:47Z)
- Aggregate rank
- #321 of 743 (score 48.8)
Across the 2 benchmarks where at least ten models are scored, Phi-3-medium-4k-instruct is in the top 10% on 0 and in the top half on 2.
Its cheapest listed input price, $0.170 per million tokens, is lower than that of 71% of the 427 models with a listed price.
Phi-3-medium-4k-instruct benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| GSM8K | 91.0% | #6 of 17 | HuggingFace GSM8K leaderboard | 2026-10-13 |
| MMLU-Pro | 55.7% | #68 of 136 | HuggingFace MMLU-Pro leaderboard | 2026-10-13 |
About Phi-3-medium-4k-instruct
Phi-3-medium-4k-instruct is an open-weight model that can run on your own hardware from microsoft. The cheapest API listing we track charges $0.170 per million input tokens and $0.680 per million output tokens, through azure_ai. We hold 2 benchmark results for it across 2 areas, and none is in the top tenth of its board.
Where Phi-3-medium-4k-instruct stands
On our aggregate score, Phi-3-medium-4k-instruct is #321 of 743 ranked models, around the middle of the table. Its score is 48.8; the scale is centred on 50, so it sits 1.2 points below the typical model. That beats 57% of the ranked field. Among the 4 ranked microsoft models it is #1. The next one down is Phi-3.5-mini-instruct at 46.5.
Phi-3-medium-4k-instruct, benchmark by benchmark
GSM8K puts Phi-3-medium-4k-instruct at 91.0%, #6 of 17. That beats 65% of the models tested, but 8.6 points short of mimo-v2.5-pro (99.6%). The results either side are granite-4.1-8b (92.5%) and Qwen2-72B (89.5%). It covers grade-school math word problems. The board median is 86.9%, so Phi-3-medium-4k-instruct is 4.1 points above it. Source: HuggingFace GSM8K leaderboard, captured 2026-10-13.
On MMLU-Pro, Phi-3-medium-4k-instruct reaches 55.7%. It ranks #68 of 136, ahead of 50% of the field, and 32.4 points behind Darwin-180B-RSI at 88.1%. The results either side are granite-4.1-8b (56.0%) and DeepSeek-V2-Chat (54.8%). It covers 57-subject knowledge + reasoning. The board median is 55.3%, so Phi-3-medium-4k-instruct is 0.4 points above it. Source: HuggingFace MMLU-Pro leaderboard, captured 2026-10-13.
Strengths and weaknesses
Phi-3-medium-4k-instruct has too few results on boards with ten or more models to say where it is strong or weak. Treat any single score as a data point, not a verdict.
Price and value
The median input price across the 427 priced models is $0.500 per million tokens, so Phi-3-medium-4k-instruct costs less than the typical listing. Models that score within 1.5 points or better and cost less: Qwen2.5-Coder-7B-Instruct ($0.010, 49.5), DeepSeek V4.1 Flash (high) ($0.010, 48.9), DeepSeek V4 Flash 0731 (max) ($0.016, 53.1). Similar scores cost more on qwen3-235b-a22b ($0.180) and codestral-2405 ($0.200).
Reading these numbers
The 2 scores on this page come from 2 sources: HuggingFace GSM8K leaderboard (1), HuggingFace MMLU-Pro leaderboard (1). No official leaderboard supplied any of them, so treat them as third-party measurements. With 2 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #318 Gemini 3.6 Flash (High) (48.9)
- #319 Claude Opus 4 (48.9)
- #320 Gemini 2.0 Flash (Experimental) (48.8)
- #322 Claude Code + Opus 4.8 (48.8)
- #323 glm-5.2-max (48.8)
- #324 Opus 4.5 (Thinking, 16K) (48.8)
More from microsoft
- Phi-3.5-mini-instruct (#507, 46.5)
- Phi-3-mini-128k-instruct (#527, 46.3)
- Phi-4-mini-instruct (#721, 41.8)
Compare it on the live leaderboard → · How the aggregate score works