agentboards.org
Compare/Legion vs Sim

LegionvsSim

Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.

Legion
Software Mansion · Agent framework
OSS
Panel
6.5
3 spec wins
Reliability
6.0
Usefulness
6.2
Cost
8.0
Longevity
5.8

“It writes Lua to get work done in Elixir, which is two functional languages agreeing that a third one should take the blame.”

Sim
Sim · Agent framework
OSSMCP
Panel
6.5
4 spec wins
Reliability
6.2
Usefulness
6.7
Cost
7.0
Longevity
6.2

“Formerly Sim Studio, now Sim, saving you six characters and creating no confusion whatsoever.”

Spec by spec

SpecLegionSim
Architecture
CategoryAgent frameworkAgent framework
Runslocalcloud, local
Platformsmacos, linuxmacos, linux, windows, web
Context windownot documentednot documented
Protocols
MCP clientNoYes
MCP serverNoNo
Capabilities
Runs terminal commandsNoNo
Multi-file editsYesNo
Git operationsNoNo
Browser controlNoNo
Sandboxed executionYesLegion runs generated Lua or Elixir in its own sandbox; the README does not describe a container runtime. No
Multi-agent orchestrationNoYes
Headless / CI modeNoNo
Models
Backbonemultiple providersany
Bring your own modelYesYes
Local modelsNoYes
Cost
Pricing modelbyokmixed
Starts at$0/mo$25/mo
Free tierYesYes
Bring your own keyYesYes
Openness
Open sourceYesYes
LicenseMITApache-2.0
GitHub stars22029,765

Which one would each critic pick

CriticLegionSimPick
El Juez——not enough reviews
El Amigo6.86.5Legion — Pick Legion if you write Elixir and want the agent inside the application rather than beside it; pick a framework in Python if the rest of your stack already lives there.
El Crítico6.05.8Legion — The generated snippet runs in the project's own sandbox, and the README does not describe a container runtime, so the isolation boundary is whatever the host process could already reach.
El Profesor7.86.8Legion — One evaluation can filter, branch and loop, so a sequence that costs a tool-calling agent a model round trip per step collapses into a single call. The claim is structural and checkable.
La Inversora6.36.5Sim — A thousand integrations plus self-hosting bundled into custom enterprise terms is the standard open-core ladder, and the enterprise tier is where this actually gets sold.
La Jefa5.55.8Sim — Twenty-five dollars per user is $1,500 a month for sixty, and support is a community repository, with directory integration only appearing in negotiated terms.
El Hacker6.87.8Sim — Apache-2.0, npx sim-setup puts it on my own box with Docker, local model endpoints work, and it speaks MCP as a client, which is the whole checklist.

Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.

Want a third column? The compare tool handles any two agents. Three-way comparisons are on the roadmap once the spec rows are all verified.