Gemini 2.5 Pro.
Google's frontier reasoning model with native 1M-token context.
No self-host path — closed weights.
Gemini 2.5 Pro's weights aren't published. Use it via the access providers below.
Cheapest hosted endpoints.
What it's best at.
Speed across providers.
Tokens/sec and time-to-first-token measured against the same prompt template on each provider's API.
| Provider | Tokens/sec | TTFT | Total |
|---|---|---|---|
| OpenRouter | 81.2 | 5911 ms | 6107 ms |
Variants in the Gemini family.
Frequently asked.
How do I run Gemini 2.5 Pro?
Where can I access Gemini 2.5 Pro?
How much does it cost to run Gemini 2.5 Pro?
Is Gemini 2.5 Pro open-source or proprietary?
What it costs per month across providers.
Estimate your monthly bill for Gemini 2.5 Pro across every host that publishes per-token pricing. Slide your token volumes; the chart + table re-rank cheapest-first.
Cheapest provider on the left.
Total monthly cost — input + output tokens combined.
Bill breakdown.
What it's best at.
Scores normalised against benchmark ceilings (100 = perfect). Coloured by tier — coral 80+ frontier, lavender 65+ strong, sage 50+ solid, slate below.
Published scores.
| Benchmark | Score | Source |
|---|---|---|
| GPQA | 64.0 | official ↗ |
| MATH | 92.0 | official ↗ |
| MMLU | 85.0 | official ↗ |
| MMMU | 81.7 | official ↗ |
| MMLU-Pro | 82.0 | official ↗ |
| HumanEval | 87.6 | official ↗ |
| SWE-bench | 63.8 | official ↗ |
Independent rankings.
About Gemini 2.5 Pro.
Gemini 2.5 Pro is Google's frontier model — multimodal (text, image, audio, video, code) with a native 1M-token context window (2M experimentally). Trained on TPU v5p clusters. Strong on long-context retrieval (needle-in-haystack benchmarks) and competitive on reasoning benchmarks with GPT-5 and Claude Opus. Available via Google AI Studio, Vertex AI, and the consumer Gemini app. Pricing tiered by input length — first 200K tokens at the base rate, longer at 2x. Built-in code execution + grounding with Google Search for fresh-information queries.