The Known Good Updated 2 Sep 2026 Subscribe
Anthropic: Claude Opus 4.1 vs OpenAI: GPT-5 Nano ·Anthropic vs OpenAI

Anthropic: Claude Opus 4.1 vs OpenAI: GPT-5 Nano — head-to-head comparison

Anthropic: Claude Opus 4.1 profile OpenAI: GPT-5 Nano profile All comparisons
Anthropic: Claude Opus 4.1 — Anthropic
Quality
#48 of 65
50based on 5 of 7 evaluations
Known Good Index
Speed
#331 of 339
7.5
Output tokens per second
Price
#427 of 439
$30.00
Blended per 1M tokens
In $15.00 · Out $75.00
Cache hit price
#316 of 322
$1.50
per 1M tokens (−90%)
Context
#297 of 473
200k
tokens (~300 pages)
OpenAI: GPT-5 Nano — OpenAI
Quality
#33 of 65
68based on 4 of 7 evaluations
Known Good Index
Speed
#57 of 339
93.6
Output tokens per second
Price
#60 of 439
$0.14
Blended per 1M tokens
In $0.05 · Out $0.40
Cache hit price
#19 of 322
$0.005
per 1M tokens (−90%)
Context
#186 of 473
400k
tokens (~600 pages)
Summary

On the Known Good Index, Anthropic: Claude Opus 4.1 scores 50 and OpenAI: GPT-5 Nano scores 68 — a 18-point lead for OpenAI: GPT-5 Nano. They share 3 published evaluations, with Anthropic: Claude Opus 4.1 ahead on 2 and OpenAI: GPT-5 Nano ahead on 1.

Blended pricing is $30.00 per 1M tokens for Anthropic: Claude Opus 4.1 and $0.14 for OpenAI: GPT-5 Nano — OpenAI: GPT-5 Nano is about 218.2× cheaper.

Throughput is 8 tokens per second for Anthropic: Claude Opus 4.1 and 94 for OpenAI: GPT-5 Nano, with time to first token at 2.7s and 2.4s. Anthropic: Claude Opus 4.1 accepts 200k tokens of context against 400k.

Anthropic: Claude Opus 4.1 is proprietary; OpenAI: GPT-5 Nano is proprietary. 5 tracked hosts serve one or both, and 1 serve both.

Generated from tracked data on 3 Sep 2026 · how these comparisons are computed →

At a glance
MetricAnthropic: Claude Opus 4.1OpenAI: GPT-5 Nano
Known Good Index 50 based on 5 of 7 evaluations 68 based on 4 of 7 evaluations
Blended $/1M $30.00 $0.14
Output tok/s 8 94
Context window 200k 400k
Providers 4 2
Jump to section

Quality

Known Good Index and the capability indices, side by side

Index scores

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

MetricAnthropic: Claude Opus 4.1OpenAI: GPT-5 Nano
Known Good Index 50 68
Coding Index 62 —
Mathematics Index — 89
Reasoning Index 43 —
Openness Index 5 5
The Known Good

Quality Breakdown

Every evaluation both models report a published score for

Evaluations both models report

Bar = Anthropic: Claude Opus 4.1 · marker = OpenAI: GPT-5 Nano · scores ingested from Epoch AI (CC-BY 4.0)

GPQA Diamond73 / 69
Mock AIME 2024–2540 / 81
Terminal-Bench38 / 22

Right-hand figures read Anthropic: Claude Opus 4.1 / OpenAI: GPT-5 Nano. Anthropic: Claude Opus 4.1 leads on 2 of 3, OpenAI: GPT-5 Nano on 1.

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.

MetricAnthropic: Claude Opus 4.1OpenAI: GPT-5 Nano
Text Arena 1,418 —
Text Arena (style controlled) 1,448 —
The Known Good

Price & Cost

Published API pricing by token type

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.

MetricAnthropic: Claude Opus 4.1OpenAI: GPT-5 Nano
Input $/1M $15.00 $0.05
Output $/1M $75.00 $0.40
Blended $/1M $30.00 $0.14
Cache read $/1M $1.500 $0.005
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.

MetricAnthropic: Claude Opus 4.1OpenAI: GPT-5 Nano
Context window 200k 400k
Max output tokens 32k 128k
The Known Good

Speed & Latency

Mean throughput and time to first token across hosting providers, over successful measurements from the last 48 hours

Throughput and latency

Means across all tracked providers over successful measurements from the last 48 hours, derived from OpenRouter endpoint telemetry.

MetricAnthropic: Claude Opus 4.1OpenAI: GPT-5 Nano
Output tokens/s 8 94
Time to first token 2.7s 2.4s
The Known Good

Providers

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

5 hosts · 1 serve both
Full provider benchmarks →
Provider Anthropic: Claude Opus 4.1 blendedAnthropic: Claude Opus 4.1 tok/s OpenAI: GPT-5 Nano blendedOpenAI: GPT-5 Nano tok/s
Amazon Bedrock · first party $30.00 11 — —
Google · first party $30.00 3 — —
Anthropic · first party $30.00 — — —
Azure · first party $30.00 — $0.14 124
OpenAI · first party — — $0.07 126

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 →