Skip to main content

Claude Opus 5.5vs.Gemini 4 Carbon

Which one is better?

Level in the overall index: the gap is too small for a clear winner.

IntelligenceOverall index, 0 to 100 · higher is better
96

Claude Opus 5.5

Rank 1 to 2

96

Gemini 4 Carbon

Rank 1 to 5

SpeedTokens per second · higher is better
98

Claude Opus 5.5

fairly fast

100

Gemini 4 Carbon

fairly fast

PriceDollars per 1M tokens · lower is better
$8

Claude Opus 5.5

expensive

n/a

Gemini 4 Carbon

Claude Opus 5.5

Subscription

At Anthropic from €21.42 a month (Claude Pro). Through other providers from ≈ €6.36 a month (ElevenLabs Starter, with limited credits). 70 plans in Germany.

Via API

$4 input, $20 output per 1M tokens, unchanged since Sep 24, 2026.

Gemini 4 Carbon

Subscription

Subscriptions in Germany: n/a

Via API

n/a (no API price known).

Overall index with data through Oct 29, 2026 · Prices from Oct 29, 2026 · Speed: measured by Artificial Analysis · Subscriptions as of Oct 24, 2026 · Subscription prices incl. VAT

Where is the difference?

Values from 0 to 100 per task area. We only name a winner when the gap is large enough and there are enough tests.

Opus 5.5 ahead: 1Gemini 4 Carbon ahead: 1level: 4too little data: 4of 10 areas

Opus 5.5Gemini 4 Carbon0 to 100, higher is better
Areas in the overall index
CodingOpus 5.5 aheadOpus 5.5: 98Gemini 4 Carbon: 92
ReasoninglevelOpus 5.5: 95Gemini 4 Carbon: 96
Mathematicsonly Opus 5.5 has a scoreOpus 5.5: above 99Gemini 4 Carbon: not enough data
KnowledgeGemini 4 Carbon aheadOpus 5.5: above 99Gemini 4 Carbon: above 99
AgentslevelOpus 5.5: 86Gemini 4 Carbon: 88
Generalonly Opus 5.5 has a scoreOpus 5.5: 94Gemini 4 Carbon: not enough data
Multimodalonly Opus 5.5 has a scoreOpus 5.5: 93Gemini 4 Carbon: not enough data
Specialist fields
FinancelevelOpus 5.5: 92Gemini 4 Carbon: 94
LegallevelOpus 5.5: 92Gemini 4 Carbon: 94
Healthonly Opus 5.5 has a scoreOpus 5.5: 91Gemini 4 Carbon: not enough data
Show 4 more areas that do not count toward the overall index
Opus 5.5Gemini 4 Carbon0 to 100, higher is better
Cybersecurityonly Opus 5.5 has a scoreOpus 5.5: 84Gemini 4 Carbon: not enough data
Searchonly Opus 5.5 has a scoreOpus 5.5: 84Gemini 4 Carbon: not enough data
Productivityonly Opus 5.5 has a scoreOpus 5.5: 95Gemini 4 Carbon: not enough data
Educationonly Opus 5.5 has a scoreOpus 5.5: 97Gemini 4 Carbon: not enough data

See for yourself: same task, one attempt

The same prompt for both models. We show the latest attempt without fixes; it does not count toward the overall index.

Pelican on a bicycle

A community classic for free-form SVG drawing.

Claude Opus 5.5

Claude Opus 5.5: Pelican on a bicycle
A community classic for free-form SVG drawing.

Reasoning (requested): high

Current attempt
Run date
Oct 1, 2026
API provider
openrouter
Requested API model
anthropic/claude-opus-5.5
API model reported in response
anthropic/claude-opus-5.5
Reasoning setting
high
Pinned endpoint
Anthropic | anthropic/claude-opus-5.5-20260921
Temperature
Provider default
Token limit including reasoning
8,192 tokens
Test protocol
community-visual-tests-v1
Technical checks
Passed, not a quality rating
Response time
33.2 s

Visually reviewed. This individual test does not establish overall model quality.

Gemini 4 Carbon

Not tested yet

Prompt and test conditions

Generate an SVG of a pelican riding a bicycle

Token limit including reasoning: 8,192 tokens

Task origin (Simon Willison)
SVG short test

There is no shared test round for this selection. Results from different rounds are not mixed.

All data side by side

The better value is bold. For the overall index only when the gap is clear.

MetricClaude Opus 5.5Gemini 4 Carbon
Intelligence
Overall index0 to 100, higher is better9696
RankRank 1 to 2 of 95Rank 1 to 5 of 95
Price per 1M tokens
Inputwhat you send$4n/a
Outputwhat the model writes$20n/a
Speed
Tokens per second98100
Technical
Context windowhow much text the model reads at once1 M tokens≈ 1,500 A4 pages1 M tokens≈ 1,500 A4 pages
Knowledge cutoffJune 2026n/a
ReleasedSep 22, 2026n/a
AvailabilityActiveLimited access
Open weightsnono
  • Gemini 4 Carbon, Announced introductory pricing: $2 input and $10 output per 1M tokens. Cached input receives a 95% discount, giving $0.10 per 1M tokens. This announcement does not establish an already available public API. Source
  • Gemini 4 Carbon, Announced pricing after the introductory period: Afterward, $4 input and $20 output per 1M tokens. Google has not specified when the introductory period ends. Source
  • Gemini 4 Carbon, Artificial Analysis pricing assumptions: Artificial Analysis calculates test costs using $4 input, $20 output, and $0.20 cached input per 1M tokens. These are the evaluator’s pricing assumptions, not confirmation of a publicly available API tariff. Source

Prices without caching or batch discounts · Page count as a rule of thumb from Artificial Analysis

Shared benchmarks

Single results from the same benchmark version, task and metric. A value is bold only with the same test setup and a clear gap.

A: Claude Opus 5.5B: Gemini 4 Carbon

Vals IndexOverall · higher is betterTest setup not documented as identicalClaude Opus 5.5: 70.2%Gemini 4 Carbon: 70.4%Vals AIAs of 10/27/2026
Vals IndexCode Migration v1.1 · higher is betterTest setup not documented as identicalClaude Opus 5.5: 80.5%Gemini 4 Carbon: 80.1%Vals AIAs of 10/27/2026
Vals IndexEMB v1.1 · higher is betterTest setup not documented as identicalClaude Opus 5.5: 75.8%Gemini 4 Carbon: 76.5%Vals AIAs of 10/27/2026
Vals IndexFinance Agent v2.1 · higher is betterTest setup not documented as identicalClaude Opus 5.5: 67.8%Gemini 4 Carbon: 67.2%Vals AIAs of 10/27/2026
Vals IndexHLAB-Vals-1.1 · higher is betterTest setup not documented as identicalClaude Opus 5.5: 9.2%Gemini 4 Carbon: 17.5%Vals AIAs of 10/27/2026
Vals IndexLegal Research · higher is betterTest setup not documented as identicalClaude Opus 5.5: 50.5%Gemini 4 Carbon: 55.8%Vals AIAs of 10/27/2026
Vals IndexTax Agent · higher is betterTest setup not documented as identicalClaude Opus 5.5: 45.6%Gemini 4 Carbon: 48.9%Vals AIAs of 10/27/2026
Vals IndexTerminal-Bench 4.0 · higher is betterTest setup not documented as identicalClaude Opus 5.5: 65.2%Gemini 4 Carbon: 63.1%Vals AIAs of 10/27/2026
Vals IndexVibe Code Bench v1.1 · higher is betterTest setup not documented as identicalClaude Opus 5.5: 90.3%Gemini 4 Carbon: 89.1%Vals AIAs of 10/27/2026

Compare another matchup

More matchups

Related LLM comparisons

Compare Claude Opus 5.5 and Gemini 4 Carbon with other leading models.

View the complete LLM comparison