Gemini 3.1 Pro PreviewvsClaude Opus 4 8 Thinking

Gemini 3.1 Pro Preview leads the overall Podium Score by 16.8 points (90.8 vs 74.0).Category wins: Gemini 3.1 Pro Preview 0 — 0 Claude Opus 4 8 Thinking.Gemini 3.1 Pro Preview is the cheaper pick ($5/M vs $15/M per 1M output tokens). Gemini 3.1 Pro Preview is faster (105 t/s vs 50 t/s).

Gemini 3.1 Pro PreviewDrag slider or hover to compareClaude Opus 4 8 Thinking
Claude Opus 4 8 Thinking
74.0
Coding: — · Speed: 50 t/s
Output Price: $15/M
Gemini 3.1 Pro Preview
90.8
Coding: 85.5 · Speed: 105 t/s
Output Price: $5/M
Quick VerdictAggregated 2026 Benchmark Analysis

Gemini 3.1 Pro Preview is rated higher overall with a Podium Score of 90.8 (vs 74.0 for Claude Opus 4 8 Thinking). Both models show closely matched capabilities across domain benchmarks.

Coding & Engineering
Comparable
Inference Speed
Gemini 3.1 Pro Preview (105 t/s)
Cost Efficiency
Gemini 3.1 Pro Preview ($5/M)
API Price / 1M (in · out)
Gemini 3.1 Pro $1.25/M · $5/M — Claude Opus 4 8 T $3/M · $15/M

Choose Gemini 3.1 Pro Preview for cost-sensitive workloads, and Claude Opus 4 8 Thinking for general-purpose workloads.

MetricGemini 3.1 Pro PreviewClaude Opus 4 8 Thinking
Podium Score90.8 ▲74.0
Arena Elo1465—
Intelligence Index91.0—
Coding85.5—
Math89.0—
Reasoning91.0—
Agentic82.0—
Knowledge88.0—
Multimodal94.0—
Long-context96.0—
Output speed105 t/s ▲50 t/s
Output price$5/M ▲$15/M
Context window1000K ▲128K
0 / 4