The Known Good Updated 25 Jul 2026
gpt-5.5-pre-release vs OpenAI: GPT-4.1 ·OpenAI vs OpenAI

gpt-5.5-pre-release vs OpenAI: GPT-4.1 — head-to-head comparison

gpt-5.5-pre-release profile OpenAI: GPT-4.1 profile All comparisons
gpt-5.5-pre-release — OpenAI
Quality
#1 of 61
98based 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-4.1 — OpenAI
Quality
#49 of 61
36based on 4 of 6 evaluations
Known Good Index
Speed
#236 of 316
33.2
Output tokens per second
Price
#268 of 340
$3.50
Blended per 1M tokens
In $2.00 · Out $8.00
Cache hit price
#174 of 206
$0.50
per 1M tokens (−75%)
Context
#49 of 345
1.05M
tokens (~1,600 pages)
Summary

On the Known Good Index, gpt-5.5-pre-release scores 98 and OpenAI: GPT-4.1 scores 36 — a 62-point lead for gpt-5.5-pre-release. They share 3 published evaluations, with gpt-5.5-pre-release ahead on 3 and OpenAI: GPT-4.1 ahead on 0.

Published pricing is tracked for OpenAI: GPT-4.1 only, at $3.50 per 1M tokens blended.

gpt-5.5-pre-release is proprietary; OpenAI: GPT-4.1 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
Metricgpt-5.5-pre-releaseOpenAI: GPT-4.1
Known Good Index 98 based on 3 of 6 evaluations 36 based on 4 of 6 evaluations
Blended $/1M — $3.50
Output tok/s — 33
Context window — 1.05M
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.

Metricgpt-5.5-pre-releaseOpenAI: GPT-4.1
Known Good Index 98 36
Mathematics Index 100 61
Science Index 99 66
Reasoning Index — 36
Agentic Index 94 33
Openness Index 5 5
The Known Good

Quality Breakdown

Every evaluation both models report a published score for

Evaluations both models report

Bar = gpt-5.5-pre-release · marker = OpenAI: GPT-4.1 · scores ingested from Epoch AI (CC-BY 4.0)

GPQA Diamond94 / 67
MATH / AIME100 / 38
SWE-bench Verified81 / 49

Right-hand figures read gpt-5.5-pre-release / OpenAI: GPT-4.1. gpt-5.5-pre-release leads on 3 of 3, OpenAI: GPT-4.1 on 0.

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.

Neither model currently appears in an arena we ingest.

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.

Metricgpt-5.5-pre-releaseOpenAI: GPT-4.1
Input $/1M — $2.00
Output $/1M — $8.00
Blended $/1M — $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.

Metricgpt-5.5-pre-releaseOpenAI: GPT-4.1
Context window — 1.05M
Max output tokens — 33k
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.

Metricgpt-5.5-pre-releaseOpenAI: GPT-4.1
Output tokens/s — 33
Time to first token — 0.8s
The Known Good

Providers

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

2 hosts
Full provider benchmarks →
Provider gpt-5.5-pre-release blendedgpt-5.5-pre-release tok/s OpenAI: GPT-4.1 blendedOpenAI: GPT-4.1 tok/s
OpenAI · first party — — $3.50 59
Azure · first party — — $3.50 42

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) Full attribution →