The Known Good Updated 25 Jul 2026
Microsoft: Phi 4 vs OpenAI: o3 ·Microsoft vs OpenAI

Microsoft: Phi 4 vs OpenAI: o3 — head-to-head comparison

Microsoft: Phi 4 profile OpenAI: o3 profile All comparisons
Microsoft: Phi 4 — Microsoft
Quality
#50 of 61
33based on 3 of 6 evaluations
Known Good Index
Speed
#142 of 316
56.1
Output tokens per second
Price
#40 of 340
$0.09
Blended per 1M tokens
In $0.07 · Out $0.14
Cache hit price
unranked
—
per 1M tokens
Context
#334 of 345
16k
tokens (~25 pages)
OpenAI: o3 — OpenAI
Quality
#28 of 61
67based on 4 of 6 evaluations
Known Good Index
Speed
#70 of 316
88.5
Output tokens per second
Price
#269 of 340
$3.50
Blended per 1M tokens
In $2.00 · Out $8.00
Cache hit price
#175 of 206
$0.50
per 1M tokens (−75%)
Context
#203 of 345
200k
tokens (~300 pages)
Summary

On the Known Good Index, Microsoft: Phi 4 scores 33 and OpenAI: o3 scores 67 — a 34-point lead for OpenAI: o3. They share 3 published evaluations, with Microsoft: Phi 4 ahead on 0 and OpenAI: o3 ahead on 3.

Blended pricing is $0.09 per 1M tokens for Microsoft: Phi 4 and $3.50 for OpenAI: o3 — Microsoft: Phi 4 is about 40.0× cheaper.

Throughput is 56 tokens per second for Microsoft: Phi 4 and 88 for OpenAI: o3, with time to first token at 0.2s and 2.7s. Microsoft: Phi 4 accepts 16k tokens of context against 200k.

Microsoft: Phi 4 is open weights under Open weights (unrestricted); OpenAI: o3 is proprietary. 2 tracked hosts serve one or both.

Generated from tracked data on 25 Jul 2026 · how these comparisons are computed →

At a glance
MetricMicrosoft: Phi 4OpenAI: o3
Known Good Index 33 based on 3 of 6 evaluations 67 based on 4 of 6 evaluations
Blended $/1M $0.09 $3.50
Output tok/s 56 88
Context window 16k 200k
Providers 1 1
Jump to section

Quality

Known Good Index and the capability indices, side by side

IndicesComponents
Index scores

Each index is min-max normalised across tracked models and scaled 0–100 from published evaluations. We do not run these evaluations.

MetricMicrosoft: Phi 4OpenAI: o3
Known Good Index 33 67
Mathematics Index 39 92
Science Index 53 84
Reasoning Index 47 62
Agentic Index — 60
Openness Index 60 5
The Known Good

Quality Breakdown

Every evaluation both models report a published score for

Evaluations both models report

Bar = Microsoft: Phi 4 · marker = OpenAI: o3 · scores ingested from Epoch AI (CC-BY 4.0)

GPQA Diamond56 / 82
MATH / AIME14 / 84
MATH Level 565 / 98

Right-hand figures read Microsoft: Phi 4 / OpenAI: o3. Microsoft: Phi 4 leads on 0 of 3, OpenAI: o3 on 3.

The Known Good

Arena Elo

Human preference rating from blind pairwise votes

Arena ratings

Elo from blind pairwise votes, ingested from Arena. Higher is better.

MetricMicrosoft: Phi 4OpenAI: o3
Text Arena 1,217 —
Text Arena (style controlled) 1,256 —
The Known Good

Price & Cost

Published API pricing by token type

Per token typeCache discount
Published pricing

USD per 1M tokens. Blended is a 3:1 input:output mix. Live from OpenRouter, cross-checked against models.dev. Lower is better.

MetricMicrosoft: Phi 4OpenAI: o3
Input $/1M $0.07 $2.00
Output $/1M $0.14 $8.00
Blended $/1M $0.09 $3.50
Cache read $/1M — $0.500
The Known Good
We publish blended price rather than cost-per-task. Cost-per-task requires token counts from an evaluation suite we do not run. Blended price is computed from published rates and is fully reproducible — see the pricing methodology.

Context Window

Maximum input accepted and maximum output emitted

Context and output limits

Provider-declared limits, from metadata via models.dev and OpenRouter. Higher is better.

MetricMicrosoft: Phi 4OpenAI: o3
Context window 16k 200k
Max output tokens 16k 100k
The Known Good

Speed & Latency

Median throughput and time to first token across hosting providers

Output speedTime to first token
Throughput and latency

Medians across all tracked providers, derived from OpenRouter endpoint telemetry.

MetricMicrosoft: Phi 4OpenAI: o3
Output tokens/s 56 88
Time to first token 0.2s 2.7s
The Known Good

Providers

Which hosts serve each model, and at what price and speed

2 hosts
Full provider benchmarks →
Provider Microsoft: Phi 4 blendedMicrosoft: Phi 4 tok/s OpenAI: o3 blendedOpenAI: o3 tok/s
DeepInfra $0.09 70 — —
OpenAI · first party — — $3.50 72

Per-host pricing and telemetry from OpenRouter. A dash means that host does not serve that model.

The Known Good
Sources for this page
Pricing & providers — OpenRouter Metadata — models.dev Benchmarks — Epoch AI (CC-BY 4.0)Elo — Arena Full attribution →