Selection guide
Which Google model fits the workload?
Use the Pro record as a reasoning and multimodal evaluation candidate. Use the stable Flash records for speed-oriented agent and coding workloads, then verify the exact API model, quota, modality, and current price in Google’s documentation.
API + app
Gemini 2.5 Pro
Consider for: Multimodal reasoning over large codebases, documents, video, and audio where a Pro-tier evaluation is appropriate.
Account for: Older release record with workload-dependent tiered pricing.
API
Gemini 3.5 Flash
Consider for: Stable sustained agent and coding workloads where the exact deployed model is already required.
Account for: Canonical numeric pricing is unavailable in the directory and computer use remains preview functionality.
API + app
Gemini 3.6 Flash
Consider for: Current stable Flash API and app workflows needing million-token multimodal input and lower listed prices than many frontier models.
Account for: Closed weights, undisclosed parameters, and a 65,536-token output limit.
Google model specifications and prices
| Model | Access | Released | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| Gemini 2.5 Pro | API + app | March 25, 2025 | 1.0M | $1.25 | $10.00 |
| Gemini 3.5 Flash | API | June 24, 2026 | 1.0M | Unavailable | Unavailable |
| Gemini 3.6 Flash | API + app | July 21, 2026 | 1.0M | $1.50 | $7.50 |
Prices are provider-listed USD snapshots per million tokens where represented. Caching, batch processing, long-context rules, modalities, regions, and service tiers can change effective cost.
Reported benchmark coverage
| Model | MMLU-Pro | GPQA Diamond | SWE-bench Verified | LiveCodeBench |
|---|---|---|---|---|
| Gemini 2.5 Pro | 86.2 | 84 | 63.8 | 48.1 |
| Gemini 3.5 Flash | — | — | — | — |
| Gemini 3.6 Flash | — | — | — | — |
A dash means the exact evaluation was not retained for that model. Unlike benchmark families and source settings are not merged into an overall score.
Provider and model sources
Each model keeps its own source record and limitations. Review the provider API reference for current authentication, quotas, SDK status, and pricing rules before implementation.