Content Humanizer

AI Content Humanizer is an Apify Actor that transforms AI-generated text into natural, human-sounding writing. It runs a two-layer Nvidia NIM + Mistral pipeline with automatic fallback and applies the blader/humanizer 35-pattern method to bypass AI detectors while preserving meaning 100%. Built for SEO content, marketing copy, and any AI-drafted text that needs to read like a person wrote it.

What Is AI Content Humanizer?

AI Content Humanizer is a text rewriting API that converts machine-generated content into natural, human-sounding text via a 2-layer resilient pipeline:

1. Primary: Nvidia NIM — openai/gpt-oss-20b (default, free endpoint Available), nvidia/nemotron-3-nano-omni-30b-a3b-reasoning (fallback) via https://integrate.api.nvidia.com/v1 (OpenAI-compatible, NVIDIA_API_KEY)
2. Backup: Mistral — mistral-small-latest, open-mistral-7b, mistral-tiny, open-mixtral-8x7b, mistral-medium-latest, mistral-large-latest

Nvidia NIM is primary because Mistral's free tier is heavily rate-limited (429s). If the Nvidia chain fails, the Actor falls back to Mistral (with retry + backoff); an explicit Mistral model reverses the order. You can also select a Mistral model directly.

The rewrite follows the blader/humanizer method: 35 patterns from Wikipedia's "Signs of AI writing" (inflated claims, sales language, stock AI words, em-dash abuse, chatbot artifacts, filler) with a two-pass draft-and-check rewrite.

Key Features

• No more 403s — free-tier safe default mistral-small-latest + automatic fallback chain
• Bypass AI detectors — 35-pattern humanization with a two-pass draft-and-check rewrite
• Match your voice — optional voiceSample: provide 2-3 paragraphs of your writing and the rewrite follows its rhythm and quirks
• Preserve meaning 100% — same facts, human voice, never invents names/numbers/dates/citations
• SEO / AIO / GEO ready — natural flow ranks better in search and AI answers
• Pay per success — $0.005 is charged only after a successful humanization

Supported AI Models

• Nvidia NIM: openai/gpt-oss-20b (default), nvidia/nemotron-3-nano-omni-30b-a3b-reasoning (fallback)
• Mistral (backup): mistral-small-latest, open-mistral-7b, mistral-tiny, open-mixtral-8x7b, mistral-medium-latest, mistral-large-latest

Verified 2026-09-04: NIM lists 81 models, but only openai/gpt-oss-20b (free endpoint Available) and nvidia/nemotron-3-nano-omni-30b-a3b-reasoning are entitled for a standard key — others (llama-nemotron, kimi, palmyra, mistral-large-2, ...) return 404 "Not found for account". openai/gpt-oss-120b went end-of-life 2026-09-03 (free endpoint Deprecated, API returns 410 Gone) and was removed.

Use Cases

• SEO & content marketing — publish AI-drafted blog posts and articles that read naturally and rank better
• AI-detector safety — strip the statistical fingerprints that AI detectors flag before you publish
• Brand voice — feed voiceSample so every rewrite matches your writing style, not a generic tone
• Editing & repurposing — de-robotize product descriptions, newsletters, and documentation drafted by LLMs
• Automation — run on a schedule or through the Apify API as part of a content pipeline

Input Parameters

{
  "content": "Machine learning algorithms utilize statistical methods...",
  "model": "openai/gpt-oss-20b",
  "nvidiaModel": "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning"
}

Field	Required	Description
content	yes	AI text to humanize (also accepts contents[0] or text for legacy)
model	no	Primary model. Default openai/gpt-oss-20b (Nvidia NIM). Use a mistral-* id to lead with Mistral instead
nvidiaModel	no	Override Nvidia NIM model (openai/gpt-oss-20b or nvidia/nemotron-3-nano-omni-30b-a3b-reasoning)
voiceSample	no	2-3 paragraphs of your own writing for voice matching (max 4000 chars, takes priority over style rules)

Sample Output

{
  "humanized": "You know how ML algorithms are like that super observant friend...",
  "modelUsed": "mistral-small-latest",
  "provider": "mistral",
  "method": "blader-humanizer-2.11.2",
  "duration": 2.8,
  "originalLength": 191,
  "humanizedLength": 761
}

provider is mistral or nvidia. modelUsed is the actual model that succeeded (may differ from input if fallback occurred). method tracks the humanizer skill version embedded in the system prompt.

How to Use the AI Content Humanizer

1. Open the Actor page on Apify Store and click Start Actor.
2. Paste the AI-generated text into content.
3. Optionally choose a model, override nvidiaModel, or add a voiceSample for voice matching.
4. Run the Actor and read the humanized text plus metadata from the dataset.

Environment Variables

Var	Required	Description
MISTRAL_API_KEY	At least one of Mistral/Nvidia	@MISTRAL_API_KEY secret
NVIDIA_API_KEY	At least one	@NVIDIA_API_KEY secret (nvapi-...)
NVIDIA_BASE_URL	no	Default https://integrate.api.nvidia.com/v1
MISTRAL_MODEL	no	Override default Mistral model
NVIDIA_MODEL	no	Override default Nvidia model

Free-tier tip: mistral-large-latest requires paid plan → 403 tier_not_allowed. Keep default mistral-small-latest or enable Nvidia fallback.

Processing Pipeline

1. Input → truncate to 12k chars if needed, prepend voiceSample (max 4k chars) when provided
2. Humanize with the blader/humanizer system prompt (35 AI-writing patterns, two-pass draft-and-check, no invented facts, em-dash ban)
3. Try Nvidia NIM chain first (openai/gpt-oss-20b → nvidia/nemotron-3-nano-omni-30b-a3b-reasoning, retry with backoff on 429/5xx). On failure, fall back to Mistral chain (preferred → env MISTRAL_MODEL → free-tier chain, same retry policy)
4. Charge PAY_PER_EVENT api-call $0.005 only after success
5. Push to dataset

Explicit Mistral model reverses the order (Mistral first, Nvidia backup).

Local Development

apify secrets add MISTRAL_API_KEY <key>
apify secrets add NVIDIA_API_KEY <nvapi-key>
apify run --purge

To run without the CLI, set MISTRAL_API_KEY and NVIDIA_API_KEY as environment variables and execute node src/main.js.

Tested: mistral-small-latest 98→750 (3s), mistral-large → fallback to small 191→761 (2.8s), openai/gpt-oss-120b 211→334 (2.2s), large 2471→3378 (8s).

Connect via API

cURL:

curl -X POST "https://api.apify.com/v2/acts/akash9078/ai-content-humanizer/runs" \
-H "Authorization: Bearer YOUR_API_TOKEN" \
-H "Content-Type: application/json" \
-d '{"content":"Machine learning algorithms utilize statistical methods...","model":"openai/gpt-oss-20b"}'

Python:

import requests

response = requests.post(
    "https://api.apify.com/v2/acts/akash9078/ai-content-humanizer/runs",
    headers={"Authorization": "Bearer YOUR_API_TOKEN"},
    json={"content": "Machine learning algorithms utilize statistical methods..."},
)
print(response.json())

Node.js with the Apify Client:

const { ApifyClient } = require('apify-client');

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });

const run = await client.actor('akash9078/ai-content-humanizer').call({
    content: 'Machine learning algorithms utilize statistical methods...',
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

AI Content Humanizer Pricing

Event	Price
Successful humanization (api-call)	$0.005

You only pay after a successful humanization (Nvidia NIM or Mistral).

Why Use This Instead of Prompting Your Own LLM?

Capability	AI Content Humanizer	DIY prompt
Fallback chain on 429/5xx	Built in (Nvidia → Mistral)	Manual
Free-tier safe defaults	Yes (mistral-small-latest)	Trial and error
35-pattern detector bypass	Embedded system prompt	You write the rules
Voice matching from a sample	voiceSample input	Not available
Pay only on success	$0.005 per event	Depends on provider
Infrastructure	Zero — runs on Apify	You maintain it

Limitations

• Input is truncated to 12k characters; voiceSample is capped at 4,000 characters
• mistral-large-latest requires a paid Mistral plan and returns 403 tier_not_allowed on the free tier
• Only openai/gpt-oss-20b and nvidia/nemotron-3-nano-omni-30b-a3b-reasoning are entitled on a standard Nvidia key; other catalog models return 404/410
• Runs on 256 MB of memory (Node 20, apify@3.4.4, @mistralai/mistralai@2.1.2)

Technical Specifications

• Engine: Node 20, apify@3.4.4, @mistralai/mistralai@2.1.2, native fetch for NIM
• Prompts: blader/humanizer v2.11.2 method (35 AI-writing patterns, two-pass rewrite, voice matching)
• Temperature: 0.7, top_p 0.9, max_tokens 8192
• Memory: 256 MB

Credits

• Humanization method: blader/humanizer v2.11.2 (MIT) — installed in this repo under .agents/skills/humanizer via npx skills add blader/humanizer. The actor embeds a condensed version of its 35 patterns in the system prompt (src/main.js).
• Pattern source: Wikipedia's "Signs of AI writing", maintained by WikiProject AI Cleanup.

Data Security

• No retention after processing, GDPR compliant

FAQ

Why does the Actor default to Nvidia NIM?

Mistral's free tier is heavily rate-limited (429s), so Nvidia NIM is the primary chain and Mistral serves as the automatic backup.

Which Nvidia NIM models can I use?

openai/gpt-oss-20b (default) and nvidia/nemotron-3-nano-omni-30b-a3b-reasoning. Most other catalog models return 404 "Not found for account" for a standard key.

Can I lead with Mistral instead?

Yes — pass an explicit Mistral model (e.g. mistral-small-latest) and the order reverses: Mistral first, Nvidia as backup.

Does humanizing change the facts?

No. The method preserves meaning 100% and never invents names, numbers, dates, or citations.

How do I make the output match my writing style?

Provide 2-3 paragraphs of your own writing in voiceSample (max 4,000 characters). The sample takes priority over default style rules.

Why did mistral-large-latest return 403?

mistral-large-latest requires a paid Mistral plan (403 tier_not_allowed). Keep the default mistral-small-latest or enable Nvidia fallback.

Related Actors

If you need more AI content tools from Apify, check out my other Apify Actors:

• AI Image Generator - Text to Image
• PDF Text Extractor
• Analyze Image
• Audio File Converter

---

Built and maintained by akash9078 on the Apify platform.