← Video

HiggsfieldHiggsfield

Higgsfield AI

免费+订阅Free + subscription 镜头控制Camera control 运镜预设Camera presets 图生视频Image-to-video 直连可用Direct access partial lifecycle: active 核验Checked 2026-10-22

Higgsfield 把「运镜」做成产品：预设镜头语言（dolly/crash zoom/crane 等）一键套用到生成视频—— 它卖的是「导演感」，不是通用生成器。

Higgsfield turns camera moves into a product: preset camera language (dolly/crash zoom/crane…) applied to generated video in one click — it sells 'directorial feel', not a general generator.

适合与不适合When to use it, when not to

适合：镜头语言可控的短片、动感社交素材、把运镜外包给预设的创作者。不适合：自由导演式控制、开放权重、长叙事一致性主打。

Fits: shorts with controllable camera language, dynamic social material, outsourcing camerawork to presets. Not a fit: free-form directing, open weights, long narrative consistency.

中文生态China availability 国产 / 国内可直接用 / 备案状态 · 快照domestic / direct access / CAC filing · snapshot 2026-10-22

出品属地Origin: 国外出品foreign-made　Higgsfield AI（叙事为海外团队；具体注册主体未核验到官方声明）。Higgsfield AI (overseas-team narrative; the exact registered entity was not verified).
国内使用Direct access: 大陆可直连direct OK　higgsfield.ai 大陆直连实测 200（2026-10-22）。higgsfield.ai measured 200 direct from mainland (2026-10-22).
备案状态CAC filing: 未见于备案名单not in filing list：未见于网信办公示名单（2026-10-22 检索）；该产品未面向大陆公众提供服务，备案不适用。not in the CAC public registry (searched 2026-10-22); the product does not serve the mainland public, so filing does not apply.

备案信息以网信办《生成式人工智能服务已备案信息》公示名单为准（按批次滚动更新，本站只抄编号不解读）；「大陆可直连」是本站从大陆网络直连实测的结果，不是厂商承诺。GitHub 类站点当日实测有波动，口径写在依据里。Filing status is quoted from the CAC public registry (updated batch by batch; this site copies the number and does not interpret it). "Direct access" is this site's own reachability measurement from a mainland network, not a vendor promise. GitHub-hosted sites fluctuated on the measurement day — the note says so.

固定坐标系Fixed coordinate system

8 个维度dimensions, 与同分区其他对象逐项可比each comparable item by item against others in the same section.

模型与开放条件Model & openness: 聚合+自研混合：平台内提供多家模型生成入口（官网导航所见）+ 自研线（公司名即模型品牌叙事）。具体聚合清单未逐条清点。
Aggregation + in-house hybrid: the platform offers generation entries for multiple models (as seen in site navigation) + an in-house line (the company name doubles as the model-brand narrative). The concrete aggregation list has not been inventoried item by item.
运行位置Where it runs: 网页（higgsfield.ai）；生成在厂商云端。未见官方离线权重线。
Web (higgsfield.ai); generation on the vendor's cloud. No official offline-weights line seen.
本地文件Local files: 作品存账户侧云端，生成历史在网页作品页可见；下载归本地。未见官方声明的本机数据面。
Works stored account-side in the cloud, generation history visible on the web works page; downloads go local. No officially stated local data surface.
关机后的任务Tasks after shutdown: 生成即时/排队（按档位波动，未逐条核验）；无「关机续跑」任务面（未核验到声明）。
Generation is instant/queued (varies by tier, not verified item by item); no "keep-running-after-shutdown" task surface (no claim verified).
工具与扩展Tools & extensions: 镜头控制（Camera Control / Move 系列）是它的招牌差异（官网主打叙事：预设运镜——dolly/ crash zoom/crane 等一键套用）；图生视频、特效模板线（官网导航所见）。
Camera Control (the Move series) is its signature differentiator (the site's lead narrative: preset camera moves — dolly/crash zoom/crane etc., one-click application); image-to-video, effect-template lines (as seen in site navigation).
上下文与记忆Context & memory: 以账户作品组织；跨会话记忆未核验到官方声明。运镜预设是它的「一致性」机制——可复用不是记忆。
Organized by account works; cross-session memory — no official statement verified. Camera-move presets are its "consistency" mechanism — reusable, not memory.
权限与限制Permissions & limits: 账号+订阅体系（D9：价格不采）；训练用途开关未逐条核验；免费档输出的商用与水印限制按条款（未逐条核验）。
Account + subscription system (D9: prices not collected); the training-use switch not verified line by line; commercial and watermark limits on free-tier output follow the terms (not verified line by line).
适合什么任务Best for: 适合：运镜可控的短片（预设镜头语言）、把「导演感」外包给预设的创作者、社交向动感素材。不适合：要开放权重、大陆直连敏感场景（直连 200 可用但产品面向海外）、长叙事一致性主打。
Fits: camera-controlled short films (preset camera language), creators outsourcing "director feel" to presets, dynamic social-media material. Does not fit: open weights, scenarios sensitive to mainland direct connection (direct 200 works but the product targets overseas), a long-narrative-consistency focus.

头号误解Top misconception

「镜头控制」是预设套用不是自由导演——运镜模板库的边界决定上限，别按专业摄影理解'Camera control' is preset application, not free directing — the preset library's boundary sets the ceiling
聚合面里的模型与自研线并存——「这个效果是谁的模型」按官网逐项看Aggregated models and self-developed lines coexist — 'whose model made this effect' must be read off the official site item by item
大陆直连实测 200（2026-10-22）——它是少数直连可用的海外视频站，但账号/支付面按海外口径Mainland direct access measures 200 (2026-10-22) — a rare overseas video site reachable directly, but accounts/payments run on overseas rails

未知项清单Known unknowns

注册主体与团队背景（未核验到官方声明）
自研模型与聚合模型的边界清单（未逐条清点）
训练用途开关（未逐条核验）
跨会话记忆机制（未核验到官方声明）

The registered entity and team background (no official statement verified)
The boundary list between in-house and aggregated models (not inventoried item by item)
The training-use switch (not verified line by line)
Cross-session memory mechanism (no official statement verified)

证据来源Evidence sources

判断可回到以下一手源复核。本站核验日Every judgement can be rechecked against the primary sources below. Verified on 2026-10-22，内容更新日content updated on 2026-10-22。

类型Type	名称Name	链接Link
official	Higgsfield · 官网（产品与能力叙事）Higgsfield · official site (product and capability narrative)	https://www.higgsfield.ai
official	Higgsfield · Terms of Use（官方条款，2026-10-22 直抓核验）Higgsfield · Terms of Use (official terms, direct-fetched and verified 2026-10-22)	https://www.higgsfield.ai/terms-of-use-agreement
changelog	Higgsfield · Blog（官方更新线，changelog）Higgsfield · Blog (official update line, changelog)	https://www.higgsfield.ai/blog

实测记录Measured runs

本档未跑生成质量与速度测试；结论只覆盖能力面与许可面。

实测建议（三条，都指向官方可读页没给的东西）：

运镜预设对复杂主体（多人/快速动作）的跟随质量
聚合模型清单里各家输出的风格差异
预设运镜与自定义提示的组合冲突行为

This site has not completed measured runs. The coding-oriented test protocol is at protocols/coding/_protocol.md (the coding task suite does not apply to creative products).

Suggested measurements (three, all pointing at things the official readable pages do not give):

Camera-preset tracking quality on complex subjects (multiple people / fast motion)
Style differences across vendors' outputs in the aggregated model list
Conflict behavior between preset camera moves and custom prompts