Tisane API

Tisane API is a content moderation service focused on text, with an API that returns human-readable explanations for its decisions. The RESTful API returns detailed results in milliseconds, and the product describes handling slang, misspellings, and algospeak, with analysis tailored to cases such as usernames and search queries. It can also identify topics, entities, and sentiment, and its language capabilities include translating foreign content while retaining original profanities and slurs. Gaming-adapted moderation distinguishes in-game violence references from criminal activity and can detect scammers, groomers, and trolls. Deployment options described include public cloud, on-prem installation, and an embedded in-process library. The Community plan costs nothing and includes 50,000 requests, a 10-hits-per-minute limit, and no technical support. Small costs $99 a month and includes 100,000 requests and limited technical support. Image, audio, and video moderation are not included, and there is no human review workflow. Tisane Labs is headquartered in Singapore.

Tisane API focuses on text moderation, with detailed explanations and several deployment options. Check the plan limits and support level against your expected request volume.

Tisane API may suit teams moderating text who need explainable decisions or custom policies. Its gaming-oriented analysis may also fit services that need to distinguish in-game language from criminal activity.

Tisane API is a text-moderation service for teams handling user-generated language, with particularly strong relevance to gaming communities and multilingual products. Its explainable, deterministic decisions suit developers who need to investigate flags, while its text-only scope rules out mixed-media moderation.

Overview

Tisane Labs develops natural-language processing tools for content moderation and law-enforcement use cases. The RESTful API returns detailed results in milliseconds and explains moderation decisions in human-readable terms. Deterministic outputs and traceable reasoning are practical advantages when policy teams need consistent outcomes or engineers need to debug a decision.

This is a focused text product. It supports custom policies, but not image, audio, or video moderation, and it does not provide a human-review workflow. A service that combines several media types or routes borderline cases to reviewers will be a better fit elsewhere.

Key features

Tisane targets the messy language that defeats simplistic keyword filters. It handles algospeak, slang, and misspellings, and can tailor analysis to contexts such as usernames and search queries. That makes it more suitable for community platforms than a one-size-fits-all text check.

Foreign-language handling includes translation while retaining the original profanities and slurs, helping moderators preserve the evidence behind a decision. Beyond moderation, the API detects topics, entities, and sentiment, which can support broader text classification workflows.

Gaming-specific analysis distinguishes references to in-game violence from criminal activity and can identify scammers, groomers, and trolls. For a game or gaming social layer, that contextual distinction is more useful than treating every violent term as equally threatening.

Deployment is flexible: customers can use the public cloud, install it on premises, or run Tisane Embedded as an in-process library. On-premises and air-gapped operation address stricter privacy requirements. Plugins cover Slack, Minecraft, Webex, Discord, TikTok, Instagram, and YouTube, while Zapier, Make, and n8n extend integration options.

Pricing

The freemium structure gives small projects a low-risk starting point, but throughput and support change substantially by tier.

Every paid SaaS tier publishes both a monthly request allowance and a rate limit, so projected volume and peak concurrency should be checked together. Community's 10-hit-per-minute ceiling is especially restrictive for live products.

Platforms

Tisane supports API, Linux, macOS, self-hosted, web, and Windows environments. Public-cloud use is complemented by on-premises and embedded deployment, including air-gapped installations for privacy-sensitive workloads.

Who it's for

Choose Tisane for gaming communities, social products, search or identity systems, and multilingual platforms where explainable text decisions matter. It is also a strong candidate when deterministic behavior, custom policies, or private deployment are requirements.

It is not the right single service for image, audio, or video moderation, nor for organizations that need built-in human escalation. Teams needing broad media coverage should pair it with another system or select a broader alternative.

Pros and cons

Pros

• Human-readable, deterministic explanations make moderation decisions traceable and debuggable.
• Handling of algospeak, slang, misspellings, and contextual scenarios addresses common weaknesses in basic filters.
• Gaming-specific detection separates fictional violence from criminal behavior and targets scammers, groomers, and trolls.
• Cloud, on-premises, air-gapped, and embedded options accommodate different privacy and architecture needs.

Cons

• Text-only coverage excludes image, audio, and video moderation.
• No human-review workflow means teams must build their own escalation process.
• Community and Small omit key capabilities, while meaningful support begins at Medium.
• The free tier's 50,000 requests and 10 hits per minute limit it to modest workloads.

Alternatives

For image-focused moderation, WebPurify offers paid image and live-moderation options across API, web, mobile, and self-hosted platforms. PicPurify is a freemium API and web choice with a 2,000-unit free plan and a $19-per-month Starter tier. Sightengine suits teams wanting freemium API/web moderation with a 1-request-per-second free tier and paid operation quotas.

Amazon Rekognition Content Moderation is the alternative when AWS image and video analysis, including a 12-month free tier, is the priority. For self-hosted toxic-comment classification at no license cost, choose Detoxify. Hive Moderation is a paid API/web option with model access and pay-as-you-go credits. Google Cloud Vision API fits image analysis billed per image. Llama Guard is a free, self-hosted model option.

Verdict

Tisane API is a strong choice for teams moderating complex user-generated text, especially gaming and multilingual content, because it combines contextual analysis with explainable, deterministic results and flexible deployment. Look elsewhere if your primary need is media moderation or a managed human-review queue; otherwise, select the plan whose request and rate limits match both normal volume and peak traffic.