Gemini 2.5 Pro vs Claude Sonnet 4.6
Google vs Anthropic · Prices checked 12 August 2026
The two are closer than they look: Gemini 2.5 Pro at $3.88 per million tokens blended, Claude Sonnet 4.6 at $6.60, a 41 per cent difference. A single 100K-input, 10K-output job costs $0.225 on Gemini 2.5 Pro against $0.450 on Claude Sonnet 4.6.
Of the 6 shared benchmarks, Gemini 2.5 Pro takes 4. The widest gap is on LMArena Elo, where Gemini 2.5 Pro leads by 37 points. 1 suite tied. Context is level at 1,000,000 tokens on each side. Gemini 2.5 Pro was released 2025; Claude Sonnet 4.6 Nov 2025.
Pricing and capacity
| Metric | Gemini 2.5 Pro | Claude Sonnet 4.6 | Difference |
|---|---|---|---|
| Input / 1M tokens | $1.25 | $3.00 | 2.4× |
| Output / 1M tokens | $10.00 | $15.00 | 33% |
| Blended · 70/30calculated | $3.88 | $6.60 | 41% |
| Blended · 50/50calculated | $5.63 | $9.00 | 38% |
| Blended · 80/20calculated | $3.00 | $5.40 | 44% |
| Blended · 20/80calculated | $8.25 | $12.60 | 35% |
| Job cost · 100K + 10Kcalculated | $0.225 | $0.450 | 2.0× |
| Job cost · 1M + 100Kcalculated | $2.25 | $4.50 | 2.0× |
| Context window | 1,000,000 | 1,000,000 | level |
Input and output rates are as published by each provider. Blended and job-cost rows are calculated comparison metrics, not published prices — how they are derived.
Benchmarks
| Benchmark | Gemini 2.5 Pro | Claude Sonnet 4.6 | Leader |
|---|---|---|---|
| MMLU | 89% | 86% | Gemini 2.5 Pro |
| GPQA Diamond | 84% | 75% | Gemini 2.5 Pro |
| MATH | 93% | 89% | Gemini 2.5 Pro |
| HumanEval | 92% | 92% | Level |
| SWE-bench Verified | 64% | 65% | Claude Sonnet 4.6 |
| LMArena Elo | 1389 Elo | 1352 Elo | Gemini 2.5 Pro |
Only suites both models report are listed, so a missing score is never shown as a loss. Figures are provider-published or from public leaderboards and are not independently re-run.
What each model is for
Gemini 2.5 Pro
Google · Frontier
Google's flagship Gemini. 1M-token context, native multimodal across text, image, audio and video, with thinking enabled.
- 1M context (2M in private beta)
- Native video + audio understanding
- Strong reasoning with thinking
- Tight Google ecosystem
Claude Sonnet 4.6
Anthropic · Balanced
Frontier-grade Claude at one fifth of Opus pricing. The default workhorse for most production traffic on the Anthropic API.
- Best price/performance in the Claude family
- 1M-token context
- Strong vision
- Reliable function calling
Frequently asked questions
Which is cheaper, Gemini 2.5 Pro or Claude Sonnet 4.6?
The two are closer than they look: Gemini 2.5 Pro at $3.88 per million tokens blended, Claude Sonnet 4.6 at $6.60, a 41 per cent difference. A single 100K-input, 10K-output job costs $0.225 on Gemini 2.5 Pro against $0.450 on Claude Sonnet 4.6. Blended cost weights input at 70% and output at 30%; price your own token split for an exact answer, since the cheaper model on a blended basis is not always cheaper for your workload.
Which performs better, Gemini 2.5 Pro or Claude Sonnet 4.6?
Of the 6 shared benchmarks, Gemini 2.5 Pro takes 4. The widest gap is on LMArena Elo, where Gemini 2.5 Pro leads by 37 points. 1 suite tied. These are the figures each provider or public leaderboard publishes, not results we re-run, and a benchmark average can hide the specific behaviour your workload depends on.
Which has the larger context window, Gemini 2.5 Pro or Claude Sonnet 4.6?
Context is level at 1,000,000 tokens on each side. The context window caps everything in a single request: prompt, retrieved documents, conversation history and the model’s own output all share it.
Can I compare Gemini 2.5 Pro and Claude Sonnet 4.6 on my own numbers?
Yes. Open both in the side-by-side comparison tool to see every field at once, or use the token calculator to price your actual prompt and monthly request volume against each model. Both are free and need no sign-up.
Price it for your own workload
Related matchups
Terms used on this page
Prices and scores are recomputed on every deploy from the catalogue. Prices checked 12 August 2026.