Agentic Research
Tool library · honest reviews

AI tools,
including who they are not for.

62 mainstream AI tools across ten categories. Every card states who it suits and who it does not, the billing model, supported platforms, and our four-dimension rating. 28 entries link to our own hands-on articles.

Tools
62
Categories
10
With our own tests
28
Price verified
7
Last updated
2026-09-29

Three promises this library keeps

  1. 01

    Every entry says who it is not for

    A review that lists only strengths carries no information. The exclusion list sits on the same card, so you know immediately whether it fits you.

  2. 02

    No amount shown unless verified

    AI tools reprice constantly. Amounts appear only on manually verified entries, stamped with the verification date; everything else states the billing model and links to the official page.

  3. 03

    Ratings are editorial, never faked as user scores

    The four dimensions (capability, ease of use, cost value, privacy control) are our judgement. We do not emit aggregateRating to impersonate a crowd-sourced score.

Ten categories

  • General Assistants8
  • AI Coding9
  • AI Search & Research9
  • Agent Orchestration9
  • Local Model Runtime5
  • Decks & Documents5
  • Image & Video5
  • Voice & Music3
  • Site & App Generation4
  • Finance & Research5

Use the filter bar below to narrow by category, level, learning track, free tier, and whether we have tested it ourselves.

62 tools

Ollama

Ollama

4.8/5

Run open models locally with one command, exposing an OpenAI-compatible API

Local Model RuntimeBeginner
Open sourceFree tierPrice verified 2026-09-29

Entirely free and open source; your only cost is hardware and electricity.

Good for
  • Privacy settings where data cannot leave the network
  • Driving token cost to zero at high call volume
  • Offline or unreliable-network environments
4 site articles1 scenariosOfficial siteView tool

Dify

LangGenius

4.5/5

Open-source LLMOps platform for visually orchestrating RAG and agent workflows

Agent OrchestrationIntermediate
Open sourceFree tierSee official pricing

Community edition is open source and self-hostable (free); cloud subscription and enterprise editions also exist. See the official page for current numbers.

Good for
  • Building maintainable RAG / agents without writing code
  • Enterprises needing private deployment
  • Centrally managing models, knowledge bases, and tools
4 site articles2 scenariosOfficial siteView tool

Open Design

開源社群 / 君澤智庫整理

4.5/5

Skill-catalogue-driven AI design output: decks, sites, dashboards, and finance documents

Decks & DocumentsIntermediateTested by us
Open sourceFree tierSee official pricing

Skills and design systems are open to use; cost comes from your model API. This site catalogues 118 skills and 146 design systems.

Good for
  • Consistent visual language across many output types
  • Automating finance documents (DCF, pitch books)
  • Codifying design standards into repeatable skills
6 site articles3 scenariosOfficial siteView tool
4.3/5

Answers only from the sources you upload — it barely fabricates

AI Search & ResearchBeginner
FreemiumFree tierSee official pricing

Usable free; a Plus variant ships with higher quota inside Google's paid AI plans. See the official page for current numbers.

Good for
  • Turning a pile of PDFs / annual reports into a queryable knowledge base
  • Generating audio overviews
  • Requiring every answer to cite the source passage
5 site articles2 scenariosOfficial siteView tool

n8n

n8n GmbH

4.3/5

Workflow automation billed per execution, with nodes that can run your own code

Agent OrchestrationIntermediate
FreemiumFree tierPrice verified 2026-09-29

Community edition is self-hostable (free, fair-code licence); n8n Cloud plans all include unlimited users and workflows, billed by monthly workflow executions. See the official page for current numbers.

Good for
  • Wiring existing SaaS together (hundreds of built-in nodes)
  • Automation needing schedules, retries, and failure alerts
  • Inserting LLM nodes for intelligent decisions mid-flow
5 site articles2 scenariosOfficial siteView tool

OpenClaw

開源社群

4.3/5

Open-source, skills-driven agent framework — extensively benchmarked on this site

Agent OrchestrationAdvancedTested by us
Open sourceFree tierSee official pricing

The framework is open source and free; your real cost is the LLM API you connect.

Good for
  • Codifying repetitive work into triggerable skills
  • Building your own multi-layer agent architecture
  • Full control over prompts and the tool chain
9 site articles1 scenariosOfficial siteView tool

LM Studio

LM Studio

4.3/5

GUI to download, manage, and run local models; supports GGUF and MLX

Local Model RuntimeBeginner
FreemiumFree tierSee official pricing

Free for personal use; commercial use requires a licence. See the official site for current terms.

Good for
  • Local models without touching a terminal
  • MLX acceleration on Apple Silicon
  • Comparing real performance across quantisations
4 site articles1 scenariosOfficial siteView tool

vLLM

開源社群(源於 UC Berkeley)

4.3/5

High-throughput LLM inference server — the default choice for self-hosted production

Local Model RuntimeAdvanced
Open sourceFree tierPrice verified 2026-09-29

Open source and free (Apache-2.0); cost is GPU hardware or cloud rental. This site publishes deployment tests and performance numbers.

Good for
  • Self-hosted inference shared by many users
  • Continuous batching to maximise GPU utilisation
  • An OpenAI-compatible endpoint for existing code
4 site articles2 scenariosOfficial siteView tool

ComfyUI

開源社群

4.3/5

Node-based local image-generation workflow, fully offline and customisable

Image & VideoAdvanced
Open sourceFree tierPrice verified 2026-09-29

Open source and free; cost is hardware (16GB+ VRAM or Apple Silicon unified memory recommended).

Good for
  • Reproducible, version-controllable image pipelines
  • High volume at near-zero marginal cost
  • Advanced control via LoRA training and ControlNet
1 site articles1 scenariosOfficial siteView tool

PaddleOCR

百度

4.3/5

Open-source OCR — we use it to turn HK annual reports into analysable structured text

Finance & ResearchIntermediate
Open sourceFree tierPrice verified 2026-09-29

Open source and free; cost is hardware plus your LLM API. This site has measured results on Traditional Chinese reports and tables.

Good for
  • Batch parsing annual reports and financial PDFs
  • Traditional Chinese documents with mixed tables
  • Self-hostable, offline-capable OCR pipelines
7 site articles1 scenariosOfficial siteView tool

Webb-site.com

Webb-site

4.3/5

Public database of HK shareholdings and directorships — a mandatory cross-check source in due diligence

Finance & ResearchBeginner
FreeFree tierPrice verified 2026-09-29

Free public lookup. Data compiled from official sources such as HKEX disclosures; this site documents usage and cross-verification method.

Good for
  • Looking up major and related shareholdings
  • Cross-directorship relationships
  • Cross-verifying HKEX filings during due diligence
4 site articles2 scenariosOfficial siteView tool

ChatGPT

OpenAI

4.0/5

The largest general assistant by user base, with the most complete ecosystem

General AssistantsBeginner
SubscriptionFree tierSee official pricing

Free tier available; Plus / Pro / Team / Enterprise paid tiers; separate usage-based API billing. See the official page for current numbers.

Good for
  • Everyday writing, translation, summarisation
  • Mature ecosystem and third-party integrations
  • Power-user data analysis and image work
8 site articles2 scenariosOfficial siteView tool

Claude

Anthropic

4.0/5

Strong long-document comprehension and writing quality; Projects can hold a knowledge base

General AssistantsBeginner
SubscriptionFree tierSee official pricing

Free tier available; Pro / Max / Team / Enterprise paid tiers; separate usage-based API billing. See the official page for current numbers.

Good for
  • Long-document reading and structured summarisation
  • Writing where tone and craft matter
  • Project-scoped knowledge-base Q&A
11 site articles2 scenariosOfficial siteView tool

Tavily

Tavily

4.0/5

Search API for AI applications, returning pre-digested, citable content

AI Search & ResearchBeginner
FreemiumFree tierSee official pricing

Free allowance, then usage-based; see the official site.

Good for
  • Web search for RAG and agents
  • Getting citable summaries directly
  • Fast integration without building your own scraping
2 site articlesOfficial siteView tool

llama.cpp

開源社群

4.0/5

A pure C/C++ inference engine — the layer most local solutions build on

Local Model RuntimeAdvanced
Open sourceFree tierSee official pricing

Entirely free and open source; your only cost is hardware.

Good for
  • Running models on constrained or unusual hardware
  • Understanding low-level quantisation and performance
  • The foundation for a self-built inference service
1 site articlesOfficial siteView tool

Claude Code

Anthropic

3.8/5

Terminal-based coding agent that can point at any OpenAI-compatible backend

AI CodingIntermediate
SubscriptionNo free tierSee official pricing

Billed via Claude paid plans or API usage. This site also documents driving it from DeepSeek / ModelStudio-compatible backends to cut cost.

Good for
  • Terminal workflows and Git operations
  • Whole-repo comprehension and batch edits
  • Codifying team process via skills / slash commands
9 site articles2 scenariosOfficial siteView tool

Gemini

Google

3.8/5

Very long context and deep Google Workspace integration

General AssistantsBeginner
SubscriptionFree tierSee official pricing

Free tier available; paid plans are often bundled with Google One; separate usage-based API billing. See the official page for current numbers.

Good for
  • Teams already on Gmail / Docs / Drive
  • Feeding extremely long documents in one go
  • Multimodal input (image, video, audio)
1 site articles1 scenariosOfficial siteView tool

DeepSeek

DeepSeek

3.8/5

Low-cost high-capability model, a good default backend for agents

General AssistantsBeginner
Usage-basedFree tierSee official pricing

Web app is free; API is token-metered at prices well below comparable Western models. This site publishes measured comparisons and price tables.

Good for
  • High-volume agent backends
  • Cost optimisation for code and reasoning workloads
  • Open-weight self-hosted deployments
10 site articles1 scenariosOfficial siteView tool

Perplexity

Perplexity AI

3.8/5

AI search engine whose answers carry inline source citations

AI Search & ResearchBeginner
SubscriptionFree tierSee official pricing

Free tier available; Pro / Max unlock more advanced models and research quota; enterprise plan available. See the official page for current numbers.

Good for
  • Fact-checking with traceable sources
  • Fast orientation in an unfamiliar field
  • Competitor and market scanning
2 site articles2 scenariosOfficial siteView tool

Exa

Exa

3.8/5

Search API designed for LLMs and agents, returning structured results

AI Search & ResearchIntermediate
Usage-basedNo free tierSee official pricing

Usage-based API billing with a free allowance; see the official site.

Good for
  • Agents needing programmatic search
  • Structured results that can be stored directly
  • Semantic rather than keyword matching
2 site articlesOfficial siteView tool

Search API on an independent index — privacy-oriented and programmatically callable

AI Search & ResearchBeginner
FreemiumFree tierSee official pricing

Free allowance, then usage-based; see the official site.

Good for
  • An independent index not tied to the big search ecosystems
  • Settings with higher privacy requirements
  • One leg of a multi-engine cross-verification setup
2 site articlesOfficial siteView tool
3.5/5

Build bots with no code and publish to multiple chat platforms in one click

Agent OrchestrationBeginner
FreemiumFree tierSee official pricing

Free quota plus paid plans; the China edition (扣子) and the global edition (Coze) differ in features and billing. See the official page for current numbers.

Good for
  • Building a first bot with zero programming background
  • Support bots published to messaging platforms
  • Fast pilots and idea validation
1 site articles1 scenariosOfficial siteView tool

v0

Vercel

3.5/5

Generates React + Tailwind interfaces from natural language, deployable directly

Site & App GenerationBeginner
FreemiumFree tierSee official pricing

Free quota plus Premium / Team / Enterprise tiers measured in credits. See the official page for current numbers.

Good for
  • Fast, showable web prototypes
  • Starting point for landing pages and dashboard UI
  • Output drops straight into a Next.js project
1 site articles1 scenariosOfficial siteView tool

Lovable

Lovable

3.5/5

Conversational full-stack app generation, including database and auth

Site & App GenerationBeginner
FreemiumFree tierSee official pricing

Free quota plus Pro / Teams / Enterprise tiers measured in credits. See the official page for current numbers.

Good for
  • Non-engineers shipping a working full-stack app
  • MVPs with auth and a database
  • Rapid product idea validation
1 site articles1 scenariosOfficial siteView tool

Bolt.new

StackBlitz

3.5/5

Generates, runs and deploys full-stack apps in the browser with no local environment

Site & App GenerationBeginner
FreemiumFree tierSee official pricing

Free allowance, then credit- and subscription-based; see the official site.

Good for
  • Building an app without a local dev environment
  • Prototypes that run and can be shared immediately
  • Rapid product idea validation
1 site articlesOfficial siteView tool

Midjourney

Midjourney Inc.

3.3/5

The highest-aesthetic text-to-image model, with fine style control

Image & VideoBeginner
SubscriptionNo free tierSee official pricing

Subscription only, tiered by GPU time. See the official page for current numbers.

Good for
  • Brand imagery, concept art, posters
  • Marketing assets with high aesthetic demands
  • Visually consistent series
2 site articles1 scenariosOfficial siteView tool

AlphaSense

AlphaSense

3.3/5

Market-intelligence search platform for financial and corporate research, covering broker research and expert calls

Finance & ResearchIntermediate
SubscriptionNo free tierSee official pricing

Enterprise subscription, priced on request; see the official site or sales.

Good for
  • Searching large volumes of broker research and expert calls at once
  • Institutional research teams
  • Searching specific wording and figures across documents
1 site articlesOfficial siteView tool

Wind 萬得

萬得資訊

3.0/5

The financial data terminal most used by Chinese institutions, covering prices, financials and macro data

Finance & ResearchAdvanced
SubscriptionNo free tierSee official pricing

Institutional subscription, priced on request; see official sales.

Good for
  • Complete data coverage institutional research requires
  • Wanting support and service guarantees
  • Needing an Excel plug-in and data API
1 site articlesOfficial siteView tool

Jan

開源社群

4.5/5

Open-source desktop app for local LLMs, emphasising offline use and privacy

Local Model RuntimeBeginner
Open sourceFree tierSee official pricing

Entirely free and open source; your only cost is hardware.

Good for
  • Fully offline, data never leaves the machine
  • Preferring open source you can inspect
  • Wanting a simple desktop interface

Whisper

OpenAI

4.5/5

Open-source speech recognition model, self-hostable with stable multilingual performance

Voice & MusicIntermediate
Open sourceFree tierSee official pricing

The model is open source, free and self-hostable; a hosted API bills by usage — see the official site.

Good for
  • Transcribing meetings and calls
  • Needing self-hosting so data never leaves
  • Multilingual and accented audio

Flowise

開源社群

4.3/5

Open-source drag-and-drop LLM flow builder that you can self-host

Agent OrchestrationIntermediate
Open sourceFree tierSee official pricing

Open source and self-hostable (free); a cloud offering also exists — see the official site.

Good for
  • Assembling RAG and agents visually
  • Needing private deployment
  • Rapid prototyping to validate an idea

GitHub Copilot

GitHub / Microsoft

4.0/5

The most widespread in-editor AI completion, deeply integrated with GitHub

AI CodingBeginner
SubscriptionFree tierSee official pricing

Free tier plus several paid plans and an enterprise tier. See the official page for current numbers.

Good for
  • Inline completion and the gentlest onboarding
  • Teams already living in GitHub
  • PR review assistance

豆包

字節跳動

4.0/5

ByteDance's consumer assistant with strong Chinese-language and multimodal support

General AssistantsBeginner
FreemiumFree tierSee official pricing

Personal use is largely free; enterprise and API usage billed separately — see the official site.

Good for
  • Chinese-language Q&A and writing
  • Voice interaction and multimodal input
  • Integration inside the mainland China ecosystem

通義千問

阿里巴巴

4.0/5

Chat product for Alibaba's Qwen family, whose open weights can be self-hosted

General AssistantsBeginner
FreemiumFree tierSee official pricing

Web app free; API usage-based. Open-weight model releases can be self-hosted.

Good for
  • Teams wanting open weights and self-hosting
  • Balanced Chinese and multilingual ability
  • A base model you can fine-tune

Cline

開源社群

4.0/5

Open-source agent extension for VS Code that lets you point at any model backend

AI CodingIntermediate
Open sourceFree tierSee official pricing

The extension is open source and free; your cost is the model API you connect.

Good for
  • Full control over the model backend and API keys
  • Avoiding lock-in to one vendor
  • Agentic edits inside your existing editor

Aider

開源社群

4.0/5

Open-source CLI pair programmer that turns every change into a git commit

AI CodingIntermediate
Open sourceFree tierSee official pricing

Open source and free; your cost is the model API you connect.

Good for
  • Version control and revertability matter
  • Terminal-first workflows
  • Cutting cost with open models

Cursor

Anysphere

3.8/5

AI-native code editor, strongest at multi-file changes

AI CodingBeginner
SubscriptionFree tierPrice verified 2026-09-29

Hobby is free with limited agent requests; Individual starts at USD 20/mo (Pro / Pro+ / Ultra); Teams USD 40/user/mo; Enterprise custom. Usage-based billing beyond included allowance.

Good for
  • Editing an existing large codebase
  • Cross-file refactoring
  • Asking questions about code as you write

Gamma

Gamma Tech

3.8/5

One click from outline or document to web-based presentation-ready slides

Decks & DocumentsBeginner
FreemiumFree tierSee official pricing

Free tier with limited quota; Plus / Pro plans unlock more output and remove watermarks. See the official page for current numbers.

Good for
  • You already have a text outline and need slides fast
  • No templates, no manual layout
  • Turning a long document into presentable slides

Kimi

月之暗面

3.8/5

Chinese assistant known for very long context — feed it whole document sets at once

General AssistantsBeginner
FreemiumFree tierSee official pricing

Free quota plus paid plans; see the official site.

Good for
  • Handling many long documents at once
  • Chinese long-document summarisation and Q&A
  • Skipping chunking by feeding whole books
3.8/5

Chinese AI search engine whose answers carry an outline and sources; an academic mode searches literature

AI Search & ResearchBeginner
FreemiumFree tierSee official pricing

Primarily free, with a membership tier; see the official site.

Good for
  • Deep search over Chinese-language information
  • Structured, outline-style answers
  • Chinese academic and industry material

LangGraph

LangChain

3.8/5

Orchestration library that describes agent flows as state graphs, supporting interrupts and resumption

Agent OrchestrationAdvanced
Open sourceFree tierSee official pricing

The library is open source and free; a hosted platform bills by usage — see the official site.

Good for
  • Controllable, interruptible, resumable agent flows
  • Human-in-the-loop designs
  • Orchestrating complex branching and loops

CrewAI

CrewAI

3.8/5

Framework that organises multi-agent work by role, with a relatively gentle learning curve

Agent OrchestrationIntermediate
Open sourceFree tierSee official pricing

The framework is open source and free; an enterprise platform also exists — see the official site.

Good for
  • Thinking about multi-agent work as role division
  • Faster onboarding than a low-level orchestration library
  • Tasks that split naturally into a few roles

Windsurf

Windsurf

3.5/5

AI editor built around the Cascade agentic flow

AI CodingBeginner
SubscriptionFree tierSee official pricing

Free plan plus paid tiers measured in credits. See the official page for current numbers.

Good for
  • Agentic multi-step tasks inside the editor
  • Painless migration from VS Code

Felo

Sparticle

3.5/5

AI search focused on multilingual research and deck output

AI Search & ResearchBeginner
SubscriptionFree tierSee official pricing

Free quota plus paid plans; can output decks and mind maps. See the official page for current numbers.

Good for
  • Cross-language (CN/JP/EN) research
  • Turning research directly into a deck
  • Mind-map style synthesis

Napkin AI

Napkin

3.5/5

Turns paragraphs into diagrams and flow visuals ready to drop into a deck

Decks & DocumentsBeginner
FreemiumFree tierSee official pricing

Free tier available; paid plans unlock more output and commercial licensing. See the official page for current numbers.

Good for
  • Needing one diagram to explain an abstract concept
  • Fast architecture, flow, and comparison visuals
  • Not wanting to learn diagramming software

ElevenLabs

ElevenLabs

3.5/5

Lifelike text-to-speech and voice cloning with broad multilingual support

Voice & MusicBeginner
SubscriptionFree tierSee official pricing

Free tier with a monthly character quota; Creator / Pro / Scale tiered by usage. See the official page for current numbers.

Good for
  • Audiobooks, narration, podcast production
  • Multilingual dubbing (good Traditional Chinese support)
  • Marketing assets needing a consistent brand voice

Consensus

Consensus

3.5/5

Search engine that answers from academic papers, with evidence strength attached

AI Search & ResearchBeginner
FreemiumFree tierSee official pricing

Free tier available; paid plans unlock more; see the official site.

Good for
  • Answering questions from academic evidence
  • Seeing where studies agree and disagree
  • A fast scan before writing a literature review

Zapier

Zapier

3.5/5

The most widespread SaaS automation platform, now with AI and agent capabilities

Agent OrchestrationBeginner
FreemiumFree tierSee official pricing

Free allowance plus several subscription tiers billed by task volume; see the official site.

Good for
  • Wiring existing SaaS together quickly
  • Avoiding technical detail entirely
  • Needing a large library of ready integrations

Suno

Suno

3.5/5

Generates complete songs from a text prompt, including vocals and lyrics

Voice & MusicBeginner
FreemiumFree tierSee official pricing

Free allowance; paid plans unlock commercial use and more output — see the official site.

Good for
  • Needing music for marketing and events
  • Fast full-song prototyping
  • Avoiding music-production software
3.5/5

Builds an app inside Replit from a description and deploys it, including database and hosting

Site & App GenerationBeginner
FreemiumFree tierSee official pricing

Free allowance; paid plans unlock more capability and deployment — see the official site.

Good for
  • One-stop development and deployment
  • Not wanting to manage servers and databases
  • Teaching and rapid prototyping
3.5/5

Natural-language conditional stock screening for China A-shares

Finance & ResearchBeginner
FreemiumFree tierSee official pricing

Basic queries free; advanced features come with paid plans — see the official site.

Good for
  • Describing screening conditions in plain Chinese
  • Fast A-share screening and peer comparison
  • Anyone unwilling to learn a query language

Grok

xAI

3.3/5

Assistant deeply integrated with the X platform, focused on real-time information

General AssistantsBeginner
SubscriptionNo free tierSee official pricing

Bundled with paid X plans, also sold standalone and via API; see the official site.

Good for
  • Needing real-time social information
  • Users already inside the X ecosystem
  • Conversation with fewer content restrictions

Codex

OpenAI

3.3/5

OpenAI's coding agent, able to run tasks in parallel in the cloud and return diffs

AI CodingIntermediate
SubscriptionNo free tierSee official pricing

Included with paid ChatGPT plans, plus usage-based API billing; see the official site.

Good for
  • Offloading parallelisable tasks to the cloud
  • Reviewable output in diff form
  • Teams already in the OpenAI ecosystem

Trae

字節跳動

3.3/5

ByteDance's AI IDE with a built-in agent mode and multi-model choice

AI CodingBeginner
FreemiumFree tierSee official pricing

Free tier available plus paid plans; see the official site.

Good for
  • Wanting an AI IDE with a free tier
  • Low migration cost from VS Code
  • Needing a Chinese-language interface

Elicit

Elicit

3.3/5

Assists systematic literature review, extracting fields across papers in bulk

AI Search & ResearchIntermediate
FreemiumFree tierSee official pricing

Free allowance plus paid plans; see the official site.

Good for
  • Running a systematic literature review
  • Extracting consistent fields across dozens of papers
  • Tracing every conclusion back to its paper

Manus

Manus

3.3/5

General-purpose autonomous agent product: give it a task and it plans and performs multi-step work

Agent OrchestrationBeginner
FreemiumFree tierSee official pricing

Credit-based billing with a free allowance; see the official site.

Good for
  • Handing over a complete task as one package
  • Not wanting to assemble tools and flows yourself
  • Research tasks requiring cross-site operation

Notion AI

Notion

3.3/5

AI that works inside your Notion workspace and can read your existing pages and databases

Decks & DocumentsBeginner
SubscriptionNo free tierSee official pricing

Offered as an add-on or within higher tiers; see the official site.

Good for
  • Team knowledge already lives in Notion
  • Asking and rewriting right next to your documents
  • Avoiding yet another tool

Beautiful.ai

Beautiful.ai

3.3/5

Presentation tool constrained by design rules, with AI handling content and layout

Decks & DocumentsBeginner
SubscriptionNo free tierSee official pricing

Subscription with individual and team plans; see the official site.

Good for
  • Producing visually consistent decks fast
  • Not wanting to do layout yourself
  • Teams needing unified templates

Sora

OpenAI

3.3/5

OpenAI's text-to-video and image-to-video model, strong at longer shots and consistency

Image & VideoBeginner
SubscriptionNo free tierSee official pricing

Available with higher ChatGPT tiers, plus an API; see the official site.

Good for
  • Longer, coherent video clips
  • Generating video directly from text or an image
  • Fast output for marketing and concept videos
3.3/5

Kuaishou's video generation model, noted for its Chinese ecosystem and human motion

Image & VideoBeginner
FreemiumFree tierSee official pricing

Free allowance, then credit-based; see the official site.

Good for
  • Video generation from Chinese prompts
  • Human motion and dynamic scenes
  • Working inside the Chinese-language ecosystem

Runway

Runway

3.0/5

Generative video tool aimed at film and video work, with finer editing and control

Image & VideoIntermediate
FreemiumFree tierSee official pricing

Free allowance, then credit- and subscription-based; see the official site.

Good for
  • Fine-grained shot editing and control
  • Film-grade workflows
  • Stylising existing footage

Devin

Cognition

2.8/5

Positioned as an autonomous software engineer: hand it a task and it plans and executes

AI CodingAdvanced
SubscriptionNo free tierSee official pricing

Subscription plus usage-based billing across tiers; see the official site.

Good for
  • Handing off a well-specified, self-contained task
  • Asynchronous execution that does not occupy your time
  • Highly repetitive engineering chores
MAINTAINED BY JUNZE THINK TANK

Junze Think Tank

We do investment research, corporate due diligence, and AI adoption consulting. This site is where we publish our methodology — free of charge, because a method that can be inspected is the only kind worth trusting.

Start learning