DeerFlow vs Skyvern vs AgenticSeek
Side by side, as alternatives to Manus. Rows that differ are highlighted; the strongest value in a row is marked. Unknown means no source yet.
Cost and coverage
Price
$0 software; you run it
$0 software; you run it
$0 software; you run it
See pricing page source ↗
Autonomous web browsing
No: No
Partial: Partly
Partial: Partly
Confirmed: Yes
Deep research reports
Partial: Partly
No: No
No: No
Confirmed: Yes
File and code tasks in a sandbox
Partial: Partly
No: No
Partial: Partly
Confirmed: Yes
Slides generated from a prompt
Partial: Partly
No: No
No: No
Confirmed: Yes
Scheduled agent tasks
Partial: Partly
No: No
No: No
Confirmed: Yes
Connectors to your apps
Partial: Partly
No: No
No: No
Confirmed: Yes
Jobs covered
5 of 6 jobs
1 of 6 jobs
2 of 6 jobs
All 6
Popularity and activity (from GitHub)
Stars
84,212Top
23,239
27,515
—
Gain this week
Unknown
Unknown
Unknown
—
Signal
Building history
Building history
Building history
—
Commits / 4 wk
679Top
228
16
—
Latest release
v2.1.0 · 2026-09-24
v1.0.56 · 2026-10-14Top
Unknown
—
Commits / 12 wk
1,333
716
30
—
People committing / 12 mo
98+
42+
18
—
Issues opened / closed, 30 days
359 / 335
17 / 3
0 / 0
—
Facts, with sources on each project page
Runs on my machine?Differs
Confirmed: Self-hosted with Docker (recommended) or a local dev setup
Confirmed: Yes, local server and UI via pip or Docker Compose; also hosted Skyvern Cloud
Confirmed: Yes, fully local with Docker services and a local model
Hosted by the vendor
Code licenceDiffers
Confirmed: MIT
Partial: AGPL-3.0 (copyleft); anti-bot features are cloud-only
Partial: GPL-3.0 (copyleft)
—
Model weights licenceDiffers
Confirmed: No separate model weights; you bring model API keys or host a model yourself (vLLM supported)
Confirmed: No separate model weights; you bring a cloud API key or a local Ollama model
Confirmed: No separate model weights; you choose a local model (Ollama, LM Studio) or a cloud API
—
Commercial useDiffers
Confirmed: Yes, MIT code; model and search provider terms apply separately
Partial: AGPL-3.0; offering a modified version as a network service requires sharing its source
Partial: GPL-3.0 code allows commercial use if derived work stays GPL; local model terms vary
—
Paid services needed?Differs
Partial: Needs a model (API key or self-hosted) and a web search provider such as Tavily or InfoQuest; self-hosted fetchers exist
Partial: Nothing paid with a local Ollama model; cloud LLM APIs and Skyvern Cloud (with anti-bot measures) are optional
Confirmed: None required with a local model; cloud LLM APIs are optional
Subscription
HardwareDiffers
Confirmed: Minimum 4 vCPU, 8 GB RAM, 20 GB free SSD for local use (more for Docker or shared servers); a local LLM is sized separately
Confirmed: About 4 GB RAM for 1-2 concurrent browser runs, 8 GB for 3-5, 16 GB for 6-10, 32 GB+ beyond; 2 GB minimum per Kubernetes pod
Confirmed: For local models, a GPU that can run a 14B model (Magistral, Qwen, DeepSeek); with a cloud API, no GPU needed
—
Operating systemsDiffers
Confirmed: Linux recommended for servers; macOS and Windows (Git Bash) for development or evaluation
Partial: Windows documented (needs Rust and C++ build tools for pip); Docker Compose elsewhere; macOS and Linux not stated
Confirmed: Windows, macOS and Linux with Docker Desktop or Docker Engine
—
Setup effortDiffers
Confirmed: Medium; make setup writes config, then make docker-start; needs Docker Compose v2.24+ and model and search keys
Confirmed: Easy to medium; pip install plus skyvern quickstart, or docker compose up with an LLM key in .env
Confirmed: Medium; Python 3.10, Docker Compose services (SearxNG, Redis), a local model server and .env config
—
Stars over time
Switch to Growth % to compare projects of different sizes fairly.
- DeerFlow
- Skyvern
- AgenticSeek
Tracked daily since 16 Oct 2026. Dashed history: estimated from public records of when people starred the repo. Lines connect observations only; gaps contain no measured points.
Growth % starts at each project's first observed point in the shown range. Zero-star starts are unavailable.
Notes for a video or post
A plain summary with today's date and a link to the sources. Check each project page before you publish.
Want share images and a ready video description? Open a project's creator kit.
Numbers: GitHub, fetched nightly. Facts: read from each project's own documentation, with the source on its page. How we check. bytedance/deer-flow ↗Skyvern-AI/skyvern ↗Fosowl/agenticSeek ↗