Claude Sonnet 4.6 vs Gemini 2.5 Pro, API pricing and model comparison

Compare Claude Sonnet 4.6 vs Gemini 2.5 Pro side by side by input/output token price, context window, benchmark signals, capabilities, and example monthly API cost.

Comparison matrixLLM Table

Models to compare2/4

Model 1

Select model366 models

Model 2

1.0M

Select model366 models

Anthropic: Claude Sonnet 4.6

Claude API

Input

$3.00 / 1M

Output

$15.00 / 1M

Context

Monthly cost

$750.00

Benchmark

1,466Top benchmarkLMArena Text Arena

VisionToolsReasoningLong context

Google: Gemini 2.5 Pro

Gemini API

Cheapest (this scenario)

Input

$1.25 / 1M

Output

$10.00 / 1M

Context

1.0MLargest context

Monthly cost

$425.00

Benchmark

1,448LMArena Text Arena

VisionToolsJSONReasoningAudioLong context

Anthropic: Claude Sonnet 4.6

Claude API

Google: Gemini 2.5 Pro

Gemini API

Input

$3.00 / 1M

$1.25 / 1MCheapest (this scenario)

Output

$15.00 / 1M

$10.00 / 1MCheapest (this scenario)

Context

1.0MLargest context

Benchmark

1,466Top benchmark

LMArena Text Arena

1,448

LMArena Text Arena

Monthly cost

$750.00

$425.00Cheapest (this scenario)

Capabilities

VisionToolsReasoningLong context

VisionToolsJSONReasoningAudioLong context

FAQ

Model comparison notes

What should I compare first?

Start with scenario cost, context window, benchmark signal, and required capabilities such as tools, JSON output, reasoning, vision, or audio.

Why can the cheapest model be a poor fit?

Low token price does not guarantee enough context, output limit, quality, or tool support. The comparison table keeps cost and capability signals visible together.

Are route prices directly comparable?

They are useful directional comparisons, but direct provider APIs and aggregator routes can have different terms, limits, and billing details.

Popular comparisons

Open high-intent model matchups directly with ready-made links.