Skip to content
← Interactive comparison

Current benchmark matchup

Grok 4.5 vs Kimi K3

Compare current scores, source confidence, price, context, capabilities, and focused Design Arena results for both models.

xAI

Grok 4.5

xAI
Vibe Coding Index
60.5
High confidence
Moonshot

Kimi K3

Moonshot
Vibe Coding Index
67.3
Tentative evidence
Current VCI result

Kimi K3 leads by 6.8 VCI points.

This result follows the current published source evidence and can change when a scheduled refresh finds newer measurements.

Exact score differences

Side-by-side benchmark data

Measure
Grok 4.5
Kimi K3
Vibe Coding Index
60.5
67.3
Intelligence
53.8
58.8
Coding
75.0
83.9
Agentic
45.7
50.7
Blended price
$3
$6
Context window
500,000
1,048,576

Grok 4.5 evidence

100% core coverage

Complete core scores supported by at least two current verified sources.

Frequency penaltyInclude reasoningLogprobsMax tokensPresence penaltyReasoningReasoning effortResponse format

Kimi K3 evidence

100% core coverage

A current score is published, but one or more inputs remain preliminary or launch reported.

Frequency penaltyInclude reasoningMax tokensPresence penaltyReasoningReasoning effortResponse formatStop