The Known Good Updated 25 Jul 2026
o1-mini vs OpenAI: GPT-5 ·OpenAI vs OpenAI

o1-mini vs OpenAI: GPT-5 — head-to-head comparison

o1-mini profile OpenAI: GPT-5 profile All comparisons
o1-mini — OpenAI
Quality
#39 of 61
56based on 3 of 6 evaluations
Known Good Index
Speed
unranked
—
Output tokens per second
Price
unranked
—
Blended per 1M tokens
Cache hit price
unranked
—
per 1M tokens
Context
unranked
—
tokens
OpenAI: GPT-5 — OpenAI
Quality
#21 of 61
73based on 5 of 6 evaluations
Known Good Index
Speed
#138 of 316
57.0
Output tokens per second
Price
#265 of 340
$3.44
Blended per 1M tokens
In $1.25 · Out $10.00
Cache hit price
#109 of 206
$0.12
per 1M tokens (−90%)
Context
#109 of 345
400k
tokens (~600 pages)
Summary

On the Known Good Index, o1-mini scores 56 and OpenAI: GPT-5 scores 73 — a 18-point lead for OpenAI: GPT-5. They share 3 published evaluations, with o1-mini ahead on 0 and OpenAI: GPT-5 ahead on 3.

Published pricing is tracked for OpenAI: GPT-5 only, at $3.44 per 1M tokens blended.

o1-mini is proprietary; OpenAI: GPT-5 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
Metrico1-miniOpenAI: GPT-5
Known Good Index 56 based on 3 of 6 evaluations 73 based on 5 of 6 evaluations
Blended $/1M — $3.44
Output tok/s — 57
Context window — 400k
Providers 0 2
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.

Metrico1-miniOpenAI: GPT-5
Known Good Index 56 73
Coding Index — 67
Mathematics Index 69 96
Science Index 60 90
Reasoning Index 66 71
Agentic Index — 67
Openness Index 5 5
The Known Good

Quality Breakdown

Every evaluation both models report a published score for

Evaluations both models report

Bar = o1-mini · marker = OpenAI: GPT-5 · scores ingested from Epoch AI (CC-BY 4.0)

GPQA Diamond62 / 86
MATH / AIME47 / 91
MATH Level 589 / 98

Right-hand figures read o1-mini / OpenAI: GPT-5. o1-mini leads on 0 of 3, OpenAI: GPT-5 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.

Metrico1-miniOpenAI: GPT-5
Text Arena 1,317 —
Text Arena (style controlled) 1,337 —
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.

Metrico1-miniOpenAI: GPT-5
Input $/1M — $1.25
Output $/1M — $10.00
Blended $/1M — $3.44
Cache read $/1M — $0.125
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.

Metrico1-miniOpenAI: GPT-5
Context window — 400k
Max output tokens — 128k
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.

Metrico1-miniOpenAI: GPT-5
Output tokens/s — 57
Time to first token — 2.6s
The Known Good

Providers

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

2 hosts
Full provider benchmarks →
Provider o1-mini blendedo1-mini tok/s OpenAI: GPT-5 blendedOpenAI: GPT-5 tok/s
OpenAI · first party — — $3.44 84
Azure · first party — — $3.44 40

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 →