GGoogle · Released Jun 17, 2025
Gemini 2.5 Flash
Gemini 2.5 Flash is Google's state-of-the-art workhorse model, specifically designed for advanced reasoning, coding, mathematics, and scientific tasks. It includes built-in "thinking" capabilities, enabling it to provide responses with…
- Released
- Jun 17 2025
- Context window
- 1M 66K out
- Input / 1M tokens
- $0.30 from $0.09
- Output / 1M tokens
- $2.50 from $0.71
Where to run it
31 providers compared
Cheapest: QiHang at $0.09 / $0.71 per 1M input / output tokens — 84% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| QiHang | $0.09 | $0.71 | — | 1M | 66K | — | Cheapest |
| Google AI Studio | $0.15 | $1.25 | $0.015 | 1M | 66K | 100%24h | +73% |
| Kilo Gateway | $0.15 | $1.25 | $0.015 | 1M | 66K | — | +73% |
| Poe | $0.21 | $1.80 | $0.021 | 1.07M | 66K | — | +148% |
| Jiekou.AI | $0.28 | $2.38 | $0.029 | 1M | 66K | — | +230% |
| Cortecs | $0.30 | $2.49 | $0.029 | 1M | 66K | — | +246% |
| AIHubMix | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Google AI Studio | $0.30 | $2.50 | $0.03 | 1M | 66K | 100%24h | +247% |
| Modelis | $0.30 | $2.50 | — | 1M | 66K | — | +247% |
| $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% | |
| ZenMux | $0.30 | $2.50 | $0.03 | 1.05M | 66K | — | +247% |
| Merge Gateway | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Helicone | $0.30 | $2.50 | $0.075 | 1M | 66K | — | +247% |
| OrcaRouter | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Impossibl | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| NanoGPT | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| SAP AI Core | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| 302.AI | $0.30 | $2.50 | — | 1M | 66K | — | +247% |
| Perplexity Agent | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Requesty | $0.30 | $2.50 | $0.075 | 1M | 66K | — | +247% |
| Abacus | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Databricks | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| CrossModel | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Vercel AI Gateway | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| FastRouter | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Vertex | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Cloudflare AI Gateway | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Standard Compute | $0.30 | $2.50 | $0.03 | 1M | 50K | — | +247% |
| Auriko | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| FrogBot | $0.30 | $2.50 | $0.075 | 1M | 66K | — | +247% |
| NEAR AI Cloud | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Ofox | $0.30 | $2.50 | $0.03 | 1M | 66K | — | +247% |
| Google AI Studiopriority | $0.54 | $4.50 | $0.054 | 1M | 66K | 97.5%24h | +524% |
Value
Price vs. performance
Gemini 2.5 Flash Other Google Other labs Value frontier · Terminal-Bench vs. input price per 1M tokens (log scale) · Drag to pan, pinch or Ctrl + scroll to zoom, double-click to zoom in.
Benchmarks
6 benchmark scores
| Benchmark | Score | Rank | Leader |
|---|---|---|---|
| Terminal-BenchSoftware engineering · Tasks completed with the programs in a real terminal (best agent) | 17.1% | #38of 40 | GPT-5.584.7% |
| Vending-Bench 2Agents & computer use · Run a simulated vending business for a year (final balance) | $549 | #52of 61 | GPT-6 Astra$15,515 |
| SpatialViz-BenchDesign & visual · Mental rotation, folding and other spatial puzzles | 36.9% | #4of 5 | Gemini 2.5 Pro44.7% |
| SimpleBenchReasoning & knowledge · Common-sense trick questions about space, time and people | 41.2% | #40of 58 | Claude Fable 581.9% |
| CritPtReasoning & knowledge · Open-ended, research-level physics problems | 1.1% | #81of 111 | GPT-5.6 Sol32.3% |
| LMCAReasoning & knowledge · Judge the quality of conceptual arguments like experts | 27.5% | #85of 123 | Claude Sonnet 5.568.3% |
Specs
Details
- Model ID
google/gemini-2.5-flash- Developer
- Release date
- Jun 17, 2025 ·
- Context window
- 1,048,576 tokens
- Max output
- 65,535 tokens
- API price
- $0.30 / $2.50 per 1M input / output tokens
- Providers
- 31
- Input types
- file, image, text, audio, video
- Reasoning
- Yes
- Tool calling
- Yes
- Open weights
- No
- Free tier
- No
Release date and prices from OpenRouter's catalog; providers from OpenRouter and models.dev. Buying directly from a provider can cost a little differently. Speed, latency and uptime are measured on OpenRouter only; uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.