Cheapest Gemini API
Normalize Modalities, Tools and Routes最便宜的 Gemini API：统一多模态、工具与路由口径

TL;DR

Gemini comparison boundariesGemini 比较边界

Compare the same exact model version and processing mode whenever possible. If Developer API and cloud access differ in availability, data handling, quotas or features, record those as decision dimensions.

尽可能比较相同模型版本与处理模式。如果 Developer API 和云端接入在可用性、数据处理、配额或功能上不同，应记录为决策维度。

Free, Standard, Batch, Flex, and Priority are not interchangeable labels. The free tier can have different model access, rate limits, and data-use treatment. Batch and Flex reduce price for delay-tolerant work, while Priority trades a higher rate for a different service objective. A comparison should state the exact tier and test whether its completion time and availability fit the product.

Free、Standard、Batch、Flex 与 Priority 并不是可以互换的标签。免费层可能拥有不同的模型权限、Rate Limit 与数据使用方式；Batch 和 Flex 用更低价格换取对延迟更宽容的处理方式；Priority 则以更高单价换取不同的服务目标。对比时必须写明准确 Tier，并测试其完成时间与可用性是否适合产品。

Build a representative modality mix. A text-only estimate is misleading for applications that submit audio, video or documents, use cached context, or invoke grounding and tools. Count thinking tokens in billed output where the price table says they are included, and price Google Search, Maps, file search, media generation, or other tools separately under their current rules.

建立代表性的模态组合。对于提交音频、视频或文档、使用缓存上下文、Grounding 与工具的应用，只估算文本会产生误导。定价表说明思考 Token 计入输出时必须一并统计；Google Search、Maps、File Search、媒体生成或其他工具则应按其当前规则单独计价。

Gemini access path comparisonGemini 接入路径对比

Gemini pricing dimensionsGemini 定价维度

Run the apples-to-apples check执行同口径检查

Create one workload ledger instead of separate text, image, audio, and tool spreadsheets. Each request row should retain the model and tier, input modality tokens, cached context, output and thinking tokens, grounding queries, tool fees, retries, latency, result validation, and final acceptance. That makes mixed-modality cost visible and prevents free-tier experiments from being presented as production economics.

建议建立一份统一工作负载账本，而不是把文本、图像、音频与工具拆成互不相干的表。每条请求都应保留模型与 Tier、各输入模态 Token、缓存上下文、输出与思考 Token、Grounding Query、工具费用、重试、延迟、结果验证与最终验收状态。这样才能看清混合模态成本，也能避免把免费层实验误写成生产成本。

• Record official price pages, billing tier, route, model, region, and timestamp.
• Replay the same text and multimodal workload with identical limits and acceptance criteria.
• Capture tools, grounding, cache storage and reads, retries, failures, and accepted outputs.
• Test queue time and completion behavior for Batch or Flex instead of assuming only the price changes.
• Compare the paid route separately from free-tier experiments and document the data-use difference.
• Recalculate when models, packaging, tools, quotas, regions, prompts, or terms change.

• 记录官方价格页、Billing Tier、路由、模型、区域与时间戳。
• 使用相同限制与验收标准回放一致的文本和多模态负载。
• 记录工具、Grounding、缓存存储与读取、重试、失败和有效输出。
• 实际测试 Batch 或 Flex 的排队与完成行为，不要假设只有价格发生变化。
• 将付费路由与免费层实验分开比较，并记录数据使用差异。
• 模型、包装、工具、配额、区域、Prompt 或条款变化后重新计算。

Keep modality-aware usage evidence保留模态感知用量证据

Every request emits model, route, modality details, tool events, cache fields, native usage and acceptance result. The cost engine applies a timestamped official pricing snapshot and preserves unpriced dimensions for review instead of guessing.

每个请求生成模型、路由、模态详情、工具事件、缓存字段、原生用量与验收结果。成本引擎应用带时间戳的官方价格快照，并保留无法定价的维度供审查，而不是猜测。

Keep Gemini tools and external capabilities distinct区分 Gemini 工具与外部能力

Provider-native tools and QVeris capabilities may have different pricing, identity and evidence. QVeris governs Discover → Inspect → Call access to external APIs, services and live data; join both ledgers only at the parent workflow.

供应商原生工具与 QVeris 能力可能采用不同价格、身份与证据。QVeris 治理外部 API、服务与实时数据的 Discover → Inspect → 调用；二者只在父工作流层合并账本。

FAQ

Official sources and further reading官方资料与延伸阅读