Choose a Free Minute Stock Data API
for Reliable 1-Minute Bars选择免费分钟级股票数据 API
可靠获取 1 分钟 K 线
Compare 1-minute OHLCV fields, delays, sessions, history depth, rate limits, licensing, and integration tradeoffs.
比较 1 分钟 OHLCV 字段、数据延迟、交易时段、历史数据范围、
请求频率限制、许可条款与接入方案。
What a free minute stock data API must prove免费分钟级股票数据 API 必须证明什么
A usable 1-minute feed must return completed and in-progress OHLCV bars with unambiguous timestamps, session rules, adjustment policy, market coverage, and freshness. “One minute” describes aggregation length, not real-time entitlement. Before choosing a free API, replay known trading sessions, compare bars with the provider's own interface or another licensed source, and calculate whether the quota supports your symbol count and refresh pattern.
可用的 1 分钟数据源应同时返回已完成与正在形成的 OHLCV K 线，并明确时间戳、交易时段、复权规则、市场覆盖和数据时效。“1 分钟”只描述聚合周期，不代表具备实时行情权限。选择免费 API 前，应重放一个已知交易日，把 K 线与供应商界面或另一项合规数据源逐项核对，并计算配额是否能支撑标的数量和刷新节奏。
For a 1-minute chart, request timestamped OHLCV bars. A latest-price endpoint cannot reliably reconstruct each minute’s open, high, low, close, and volume.
Confirm timezone, interval boundaries, regular versus extended sessions, delayed delivery, and whether the current bar is partial or complete.
Calculate symbols × requests per symbol × refreshes × users, then add retries and backfills. Daily allowance alone does not reveal burst constraints.
Personal research, internal analytics, public display, redistribution, and automated trading can carry different exchange and vendor permissions.
绘制 1 分钟图表时，应请求带时间戳的 OHLCV K 线。仅靠最新价无法可靠还原每分钟的开盘价、最高价、最低价、收盘价和成交量。
核对时区、周期边界、常规交易时段与盘前盘后时段、数据延迟，并确认当前 K 线仍在形成还是已经结束。
按“标的 × 每标的请求数 × 刷新次数 × 用户数”计算，再加入重试与回补。只看每日总量无法判断短时限流。
个人研究、内部分析、公开展示、数据再分发和自动交易，可能分别适用不同的交易所与供应商许可。
Fields a 1-minute stock data API should return1 分钟股票数据 API 应返回哪些字段
A usable minute bar needs more than a price. Store the provider's raw response beside your normalized record during evaluation so timestamp, adjustment, and correction problems remain diagnosable.
要构成可用的分钟 K 线，仅返回单一价格远远不够。评估期间应同时保留供应商原始响应和标准化记录，便于追查时间戳、复权和数据修订问题。
| Data数据 | Fields to expect应包含的字段 | Use用途 | Common trap常见误区 |
|---|---|---|---|
| OHLCVOHLCV | Open, high, low, close, volume开盘价、最高价、最低价、收盘价、成交量 | Candles, indicators, alertsK 线、指标与提醒 | Building bars from sparse quotes用零散报价自行拼接 K 线 |
| Bar timeK 线时间戳 | Start or end timestamp, timezone, intervalK 线开始或结束时间戳、时区、周期长度 | Align bars and fill gaps对齐 K 线并识别数据缺口 | Mixing UTC and exchange time混用 UTC 与交易所时间 |
| Session交易时段 | Venue, regular/extended flag, status交易场所、常规或盘前盘后时段标记、市场状态 | Separate premarket and after-hours区分盘前与盘后交易 | Treating overnight gaps as missing data把隔夜休市区间误判为数据缺失 |
| Adjustments复权信息 | Raw/adjusted flag, splits, dividends原始/复权标记、拆股、分红 | Research and backtesting分析研究与回测 | Combining raw and adjusted series混用原始价格序列与复权价格序列 |
Define the interval boundary precisely准确约定分钟边界
Decide whether a timestamp marks the bar start or end. For a start-stamped series, a U.S. regular-session bar at 09:30 represents trades from 09:30:00 up to—but not including—09:31:00 in the exchange timezone. Do not infer this convention from array order. Document it in the normalized schema and convert to UTC only after interpreting the provider timestamp correctly.
首先确认时间戳表示 K 线开始还是结束。若按开始时间标记，美股常规时段的 09:30 K 线通常代表交易所时区内从 09:30:00 到 09:31:00 之前的成交。不能根据数组顺序自行猜测；应把规则写进内部数据模型，在正确理解供应商时间后再统一转换成 UTC。
| Semantic check语义检查 | Question to answer必须回答的问题 | Failure if ignored忽略后的后果 |
|---|---|---|
| Partial bar未完成 K 线 | Can the latest bar change until the interval closes?最新 K 线是否会在周期结束前持续变化？ | Indicators repaint and alerts trigger on unfinished data.指标重绘，提醒在未完成数据上误触发。 |
| Empty minute空分钟 | Does the API omit no-trade intervals or emit zero-volume bars?无成交分钟会被省略，还是返回零成交量 K 线？ | Forward filling may create fictitious liquidity.错误前向填充会制造虚假流动性。 |
| Corrections数据修订 | Can completed bars be revised after late or corrected trades?迟到成交或更正成交是否会修改已完成 K 线？ | Cached history diverges from the provider's final record.缓存历史与供应商最终记录不一致。 |
| Corporate actions公司行为 | Are OHLC and volume raw, split-adjusted, or dividend-adjusted?OHLC 与成交量是原始、拆股复权还是含分红复权？ | Returns, indicators, and backtests contain artificial jumps.收益、指标和回测出现人为跳变。 |
What “free” means for minute stock data“免费”的分钟级股票数据有哪些限制
Convert the advertised quota into your own workload: symbols × refreshes × users. A generous daily allowance can still fail if the provider enforces a tight per-minute burst limit.
Free feeds may be delayed, end-of-day only, or limited outside regular sessions. Label the timestamp and delay in the UI instead of implying live execution quality.
A provider may expose daily data for free while reserving 1-minute history, adjusted bars, extended hours, or larger output windows for paid plans. Confirm the exact endpoint—not the platform's broad “free data” claim.
Free access may be limited to personal or non-display use. Confirm whether you may cache, derive, display, redistribute, or use the data in a commercial product before implementation.
可按“标的数量 × 刷新次数 × 用户数”将服务商公布的配额换算为实际请求负载。即使每日配额充足，严格的短时请求频率上限仍可能导致请求失败。
免费数据源可能提供延迟行情或仅提供日终价格，也可能无法完整覆盖盘前、盘后等非常规交易时段。界面应明确标注报价时间和延迟情况，避免让用户误以为这些数据具备支持实时交易的质量。
供应商可能免费提供日线，却把 1 分钟历史、复权 K 线、盘前盘后或较大返回窗口放在付费套餐中。应核对具体端点，而不是只看平台笼统宣称“提供免费数据”。
免费访问可能只允许个人研究或非展示用途。接入前应确认是否可以缓存、生成衍生数据、公开展示、再分发或用于商业产品。
Representative supplier paths代表性供应商路径
Availability, free plans, delays, and commercial terms change. The table describes where to investigate, not a promise that every listed endpoint is free for your use case.
可用性、免费套餐、延迟和商业条款都会变化。下表用于说明应该从哪里核对，不代表所有列出的端点都能免费满足你的场景。
| Supplier供应商 | Minute-data route分钟数据路径 | Verify before choosing选用前核对 |
|---|---|---|
| Alpha Vantage | TIME_SERIES_INTRADAY supports 1-minute OHLCV plus adjustment and extended-hours controls.TIME_SERIES_INTRADAY 支持 1 分钟 OHLCV，并提供复权与盘前盘后控制。 | Current premium status, entitlement, output window, rate and license.当前付费要求、行情权限、返回窗口、限流和许可。 |
| Twelve Data | The time_series endpoint supports a 1min interval across plan-dependent instruments and markets.time_series 端点支持 1min 周期，具体标的和市场依套餐而定。 | Exchange coverage, credits per request, delay, history, commercial rights.交易所覆盖、每次请求积分、延迟、历史范围和商业使用权。 |
| Alpaca | Historical stock bars and streaming market data integrate naturally with brokerage workflows.历史股票 K 线和流式行情适合与经纪交易工作流结合。 | Feed selection, SIP versus IEX coverage, subscription, rate, account requirements.数据源选择、SIP 与 IEX 覆盖、订阅、限流和账号要求。 |
| Polygon | Stock aggregate APIs and minute flat files serve historical and market-data workflows.股票聚合 API 与分钟级 Flat File 适合历史和行情数据流程。 | Plan access, freshness, source feed, flat-file versus REST cost, display rights.套餐权限、新鲜度、源行情、文件与 REST 成本和展示权。 |
A resilient 1-minute bar integration pattern稳健的 1 分钟 K 线集成方案
Map provider-specific responses into your own Quote, Bar, and Instrument models. Keep raw payloads for debugging, but do not spread vendor field names through product code.
Cache completed historical minute bars much longer than the bar currently forming. Use explicit freshness rules and coalesce identical requests to protect the free quota.
Reject impossible timestamps, negative prices, inconsistent currencies, duplicate bars, and unexpected timezone shifts. Record the provider and retrieval time.
Use bounded historical requests to fill completed bars, then poll or stream only the live edge. Do not repeatedly download the full intraday window for every screen refresh.
After the session, re-fetch a small overlap and upsert by instrument, interval, and bar start. This captures vendor corrections without duplicating records.
将不同数据服务商的专有响应统一映射为内部的报价（Quote）、K 线（Bar）和证券（Instrument）模型。保留原始响应数据以便排查问题，但不要让服务商的字段名散落在产品代码各处。
已完成的历史分钟 K 线可采用较长的缓存期限；当前仍在形成的 K 线则应更频繁地更新。设置明确的数据时效规则，并合并相同请求，以免过快耗尽免费配额。
拒绝包含异常时间戳、负数价格、币种不一致、重复 K 线或非预期时区偏移的数据，并记录数据服务商及数据获取时间。
使用有界历史请求填充已完成 K 线，再只轮询或订阅最新边缘。不要每次刷新页面都重复下载整段盘中历史。
交易时段结束后重新获取一小段重叠区间，并按证券、周期与 K 线开始时间执行 upsert，吸收供应商修订且不产生重复记录。
Normalized bar contract: retain instrument_id, venue, interval, bar_start_utc, OHLCV, is_complete, session, adjustment, currency, provider, source_timestamp, and retrieved_at. Never mix raw and adjusted bars in one series.
标准化 K 线契约：至少保留 instrument_id、venue、interval、bar_start_utc、OHLCV、is_complete、session、adjustment、currency、provider、source_timestamp 和 retrieved_at。同一序列中绝不能混入原始与复权 K 线。
Test the feed before trusting a chart先测试数据源，再相信图表
- Replay a normal session, an early close, a daylight-saving transition, a split date, and a low-liquidity symbol.
- Assert monotonic timestamps, fixed interval alignment, nonnegative prices and volume, OHLC relationships, currency, and duplicate policy.
- Simulate quota errors, timeouts, empty responses, malformed payloads, late bars, and corrections; verify retries do not invent or duplicate data.
- Display the source timestamp and delay separately from the application's retrieval time.
- 重放普通交易日、提前收盘日、夏令时切换、拆股日和低流动性证券。
- 断言时间戳单调、周期对齐、价格与成交量非负、OHLC 关系、币种和去重规则。
- 模拟限流、超时、空响应、格式错误、迟到 K 线和数据修订，确认重试不会创造或重复数据。
- 界面应分别显示数据源时间与应用获取时间，并明确延迟。
Find minute stock data capabilities with QVeris借助 QVeris 查找分钟级股票数据接口
A live catalog check did not surface one exact public QVeris Tool or Provider page that can be guaranteed to deliver 1-minute stock bars under every requested entitlement. The page therefore links to QVeris Docs and the QVeris Playground rather than inventing a match. Use discovery to find candidates, then inspect the selected capability's schema and cost signals before calling it.
实时检查目录后，没有发现一个可公开确认、且在各种行情权限下都保证提供 1 分钟股票 K 线的 QVeris Tool 或 Provider 页面。因此本页使用 QVeris 文档 和 QVeris Playground，不会虚构匹配项。应先通过发现获取候选，再检查所选能力的数据结构与费用信号，最后调用。
- Search for quote, OHLCV, historical price, symbol lookup, and corporate-action capabilities.
- Inspect inputs and outputs before calling a tool from your code or agent.
- Keep provider selection separate from your normalized application model.
- Use the live market data provider comparison as a shortlist, then verify official supplier terms.
- 搜索提供报价、OHLCV、历史股价、股票代码查询和公司行为数据的接口。
- 通过代码或智能体调用接口前，先检查其输入、输出和调用要求。
- 将供应商选择逻辑与应用内部的标准化数据模型解耦。
- 可先参考实时更新的行情数据供应商比较形成候选名单，再核对供应商官方条款。
After discovery, compare interval boundaries, session coverage, adjustment policy, timestamps, quotas, and licensing before choosing the minute-bar capability your application will call.
找到候选接口后，还应比较 K 线周期划分规则、交易时段覆盖范围、复权规则、时间戳、请求配额和数据使用许可，再选择最适合应用调用的分钟 K 线接口。
FAQ常见问题
Yes for prototypes and limited research, but providers often restrict symbols, intraday history, request frequency, exchange coverage, or commercial display.
Not necessarily. One-minute is the bar interval, not a freshness guarantee. Check the source timestamp and documented delay.
Not reliably from occasional snapshots. Accurate bars require the relevant trade stream or provider-built aggregates with defined venue and correction rules.
Use one explicit convention for the research question. Adjusted history supports continuity; raw bars reflect traded prices. Never merge them silently.
Charts may need days; indicators need warm-up; research and validation may require months or years. Check the endpoint's actual intraday depth.
Do not assume so. Verify freshness, source feed, reliability, exchange entitlement, automation rights, and the consequences of delay or outage.
通常可以满足原型开发和小范围研究的需要，但供应商往往会限制可查询的股票数量、盘中历史数据范围、请求频率、交易所覆盖范围或商业展示权限。
不一定。1 分钟只是 K 线周期，并不代表数据具有实时性。应核对数据源时间戳和文档注明的延迟。
零散快照无法可靠完成。准确 K 线需要相关成交数据流，或供应商按明确交易场所与修订规则生成的聚合数据。
应根据研究问题固定一种明确口径。复权历史便于保持连续，原始 K 线反映实际成交价；两者不能静默混合。
图表可能只需数日，指标需要预热窗口，研究和验证则可能需要数月或数年。应核对具体端点的盘中历史深度。
不能默认可以。必须核对新鲜度、源行情、可靠性、交易所权限、自动化使用权，以及延迟或中断的后果。
References and next steps参考资料与下一步
Alpha Vantage intraday documentation
Twelve Data API documentation
Alpaca stock market data
Polygon stock data documentation
QVeris market data comparison
QVeris Docs
QVeris Playground
Alpha Vantage 盘中文档
Twelve Data API 文档
Alpaca 股票行情文档
Polygon 股票数据文档
QVeris 行情数据比较
QVeris 文档
QVeris Playground