Head-to-head comparison
Phi 3 medium 4k instruct (Microsoft) and Phi 4 (Microsoft) compared on the ModelCap Index, API price, context window, provider availability, weight access and every public benchmark board they share. Figures are the same ones shown on the live rankings; nothing here is a hidden score.
Snapshot as of 23 September 2026
As of 23 September 2026, Phi 4 holds the stronger ModelCap Index position (#209 vs #240).
Evidence: 2 public benchmark observations across 2 boards · 2 public benchmark observations across 2 boards. Prices are the lowest listed API offer per million tokens observed on the OpenRouter catalogue.
Scores are the sources' own published figures for each model's best evaluated configuration; ModelCap never re-runs a benchmark.
Want a different pairing? Open the interactive comparison tool to swap either model for any current ranked language model.
Phi 4 ranks higher on the ModelCap Index as of 23 September 2026: #209 against #240. That is a capability ranking built from public benchmark evidence with published uncertainty; whether it is "better" for you also depends on price, context and where you can run it. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.
At least one of the two has no listed API price on ModelCap right now, so no price comparison is made.
Both publish a —-token context window.
Arena coding: Phi 3 medium 4k instruct 1231, Phi 4 1306 — Phi 4 leads.
Phi 3 medium 4k instruct: Open weights. Phi 4: Open weights. Open weights mean the checkpoint can be downloaded and self-hosted under its licence; API-only models are available solely through hosted endpoints.
ModelCap currently lists 0 API providers for Phi 3 medium 4k instruct and 1 for Phi 4, from the OpenRouter catalogue snapshot the site serves; each model page lists the providers and their prices.
Every figure comes from the sealed ModelCap dataset published 23 September 2026; the page re-renders within a minute of each data refresh, and the ModelCap Index positions are the same ones shown on the live rankings.