Install
npm install @babylonjs/core
vs Unity
vs Unreal Engine
vs Godot Engine
vs three.js
vs Bevy
vs PlayCanvas
vs PhaserTry itExperimental
See what an agent can do with Babylon.js before you ever sign up. Pick a story: recorded sessions replay real probe-harness transcripts; sandboxed self-drive sessions are designed and gated (docs/TRY-IT.md).
$curl -s https://doc.babylonjs.com/llms.txt | head -4recorded session — replayed, not liveVerified integrations
No integration evidence found in our corpus for this product yet — that means none was found, never that it doesn’t integrate.
By theme — the product's score on each story themeBy theme
Agenticness — how well agents can access and operate the productAgenticnessevidence →
How well agents can access and operate the product
Ai workflows — AI in the engine loop — agent-driven editors, copilots, codegen-friendly APIs, runtime inferenceAi workflowsevidence →
AI in the engine loop — agent-driven editors, copilots, codegen-friendly APIs, runtime inference
Asset pipeline — getting content in — automated import, formats, marketplacesAsset pipelineevidence →
Getting content in — automated import, formats, marketplaces
Automation depth — how much of the product can run unattendedAutomation depthevidence →
How much of the product can run unattended
Core engine — the engine core — rendering, 2D, physics, performance at scaleCore engineevidence →
The engine core — rendering, 2D, physics, performance at scale
Editor tooling — the editor as a product — scene tools, extensibility, team workflowsEditor toolingevidence →
The editor as a product — scene tools, extensibility, team workflows
Headless automation — the engine without a human — CLI builds, CI test runs, dedicated serversHeadless automationevidence →
The engine without a human — CLI builds, CI test runs, dedicated servers
Licensing openness — the terms you build on — licenses, royalties, source access, pricing stabilityLicensing opennessevidence →
The terms you build on — licenses, royalties, source access, pricing stability
Openness — open source, data portability, and self-hosting storiesOpennessevidence →
Open source, data portability, and self-hosting stories
Platform export — shipping everywhere — desktop, mobile, console, browser payloadsPlatform exportevidence →
Shipping everywhere — desktop, mobile, console, browser payloads
Privacy posture — data-handling and privacy storiesPrivacy postureevidence →
Data-handling and privacy stories
n/a
Scripting — writing the game — languages, visual scripting, iteration speedScriptingevidence →
Writing the game — languages, visual scripting, iteration speed
Story verdicts — every judged story with its evidenceStory verdicts
Follow the green: where the map greys out is where Babylon.js stops today. ✓ full · ~ partial · ! disputed · — none · n/a not applicable.
Agenticness — how well agents can access and operate the productAgenticness
How well agents can access and operate the product
API surface
Drive the product through a documented public API
✓8/10
unlocks → MCP server · Machine-readable spec · Official CLI
Subscribe to events via webhooks
n/an/a
Build against official SDKs
✓8/10
Issue scoped/least-privilege API credentials for an agent
n/an/a
Connect an agent via an official MCP server
—0/10
Download a machine-readable API spec (OpenAPI or equivalent)
—0/10
Rely on versioned APIs with a documented deprecation policy
~4/10
Test against a sandbox environment without touching production data
~4/10
Explore an interactive API reference with runnable examples
~6/10
Docs for agents
Point an agent at llms.txt or agent-oriented docs
✓9/10
Agentic features
Delegate tasks to a built-in AI assistant inside the product
—–
Operate the product with natural-language commands
—–
Plug MCP servers into this product so it can use their tools
n/an/a
Get AI-generated insights and suggestions from my data inside the product
—–
Set up automations that run autonomously in the background
—–
Ai workflows — AI in the engine loop — agent-driven editors, copilots, codegen-friendly APIs, runtime inferenceAi workflows
AI in the engine loop — agent-driven editors, copilots, codegen-friendly APIs, runtime inference
An agent can drive the editor and project programmatically — create scenes and nodes, edit properties, trigger builds — through a documented scriptable interface
~4/10
LLM-generated code mostly works on the first try — stable, well-documented APIs with abundant public examples the models have seen
~6/10
An official AI assistant inside the engine helps with engine tasks — generating scripts, answering API questions, scaffolding scenes or assets
—0/10
Run ML models inside the game — an official inference runtime for on-device model execution
n/an/a
Asset pipeline — getting content in — automated import, formats, marketplacesAsset pipeline
Getting content in — automated import, formats, marketplaces
Automation depth — how much of the product can run unattendedAutomation depth
How much of the product can run unattended
Automation depth
Define rules that trigger actions automatically on events
~3/10
unlocks → Versioned workflows · Autonomous automations · An official AI assistant inside the engine helps with engine tasks — generating scripts, answering API questions, scaffolding scenes or assets
Schedule recurring jobs or workflows
n/an/a
Perform bulk operations across many items at once
—–
Version, review, and roll back my automations
—–
Core engine — the engine core — rendering, 2D, physics, performance at scaleCore engine
The engine core — rendering, 2D, physics, performance at scale
2D is a first-class workflow — sprites, tilemaps, 2D physics — not a 3D afterthought
~3/10
Profile and scale — a real profiler, plus data-oriented paths (ECS, jobs, instancing) when scenes get heavy
~4/10
Built-in physics — rigid bodies, collisions, raycasts — works out of the box
!5/10
The engine ships a modern production 3D renderer — PBR materials, global illumination or baked lighting, shadows, post-processing
~4/10
Editor tooling — the editor as a product — scene tools, extensibility, team workflowsEditor tooling
The editor as a product — scene tools, extensibility, team workflows
The project format and tooling play well with version control and multi-person teams — mergeable scenes, diffable text formats, or built-in collaboration
—–
A full visual editor — viewport, inspector, prefabs/scene composition — is the primary way to build levels
~4/10
Extend and automate the editor itself — custom tools, editor scripts, plugins that manipulate scenes and assets programmatically
—0/10
Headless automation — the engine without a human — CLI builds, CI test runs, dedicated serversHeadless automation
The engine without a human — CLI builds, CI test runs, dedicated servers
Build and export the project from the command line, headless, in CI — no human clicking an editor
~5/10
Game logic runs headless on servers — a dedicated-server or server-runtime build without rendering
~6/10
Tests run headlessly — unit and integration tests of game code execute in CI against the real engine
✓7/10
Licensing openness — the terms you build on — licenses, royalties, source access, pricing stabilityLicensing openness
The terms you build on — licenses, royalties, source access, pricing stability
The project's governance and funding are transparent — a foundation, published finances, or a public roadmap I can plan against
—–
The license is permissive with no royalties or per-install fees — I keep what my game earns
✓8/10
Read and modify the full engine source when I hit a wall
✓7/10
The pricing and license terms have a track record of stability — no retroactive changes that reprice games already shipped
✓7/10
Openness — open source, data portability, and self-hosting storiesOpenness
Open source, data portability, and self-hosting stories
Platform export — shipping everywhere — desktop, mobile, console, browser payloadsPlatform export
Shipping everywhere — desktop, mobile, console, browser payloads
Privacy posture — data-handling and privacy storiesPrivacy posture
Data-handling and privacy stories
Scripting — writing the game — languages, visual scripting, iteration speedScripting
Writing the game — languages, visual scripting, iteration speed
Sorted by importance (agentic first) (high → low) · 56/56 stories · click a row’s chevron for the rationale and evidence
Drive the product through a documented public API G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 3 | full | 8/10 | Tprobed | |
Connect an agent via an official MCP server G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 3 | none | 0/10 | ||
Plug MCP servers into this product so it can use their tools G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 3 | n/a | 0/10 | ||
Delegate tasks to a built-in AI assistant inside the product G Agentic features | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 3 | none | untested | none yet | |
Point an agent at llms.txt or agent-oriented docs G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | full | 9/10 | Tprobed | |
Build against official SDKs G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | full | 8/10 | Tprobed | |
Run the product headlessly / in CI for automation G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | full | 7/10 | Tprobed | |
Explore an interactive API reference with runnable examples G Api quality | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | partial | 6/10 | Tprobed | |
Rely on versioned APIs with a documented deprecation policy G Api quality | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | partial | 4/10 | Tprobed | |
Download a machine-readable API spec (OpenAPI or equivalent) G Api quality | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | 0/10 | ||
Get AI-generated insights and suggestions from my data inside the product G Agentic features | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | untested | none yet | |
Issue scoped/least-privilege API credentials for an agent G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | n/a | untested | none yet | |
Operate the product with natural-language commands G Agentic features | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | untested | none yet | |
Set up automations that run autonomously in the background G Agentic features | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | untested | none yet | |
Subscribe to events via webhooks G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | n/a | untested | none yet | |
Use an official CLI G Agent access | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 2 | none | untested | none yet | |
Test against a sandbox environment without touching production data G Api quality | ai-native user | Agenticness — how well agents can access and operate the productAgenticness | 1 | partial | 4/10 | Cclaimed | |
The license is permissive with no royalties or per-install fees — I keep what my game earns License | studio lead | Licensing openness — the terms you build on — licenses, royalties, source access, pricing stabilityLicensing openness | 3 | full | 8/10 | Tprobed | |
Self-host the core product G | ai-native user | Openness — open source, data portability, and self-hosting storiesOpenness | 3 | full | 7/10 | Tprobed | |
The primary scripting language is productive and fully exposes the engine API, with a debugger behind it Language | game developer | Scripting — writing the game — languages, visual scripting, iteration speedScripting | 3 | full | 7/10 | Xcommunity | |
Build and export the project from the command line, headless, in CI — no human clicking an editor Ci | game developer | Headless automation — the engine without a human — CLI builds, CI test runs, dedicated serversHeadless automation | 3 | partial | 5/10 | Tprobed | |
Export all of my data in open formats and leave G | ai-native user | Openness — open source, data portability, and self-hosting storiesOpenness | 3 | partial | 5/10 | Tprobed | |
One project exports to desktop, mobile, and (directly or via partners) consoles Targets | studio lead | Platform export — shipping everywhere — desktop, mobile, console, browser payloadsPlatform export | 3 | partial | 5/10 | Cclaimed | |
A full visual editor — viewport, inspector, prefabs/scene composition — is the primary way to build levels Editor | technical artist | Editor tooling — the editor as a product — scene tools, extensibility, team workflowsEditor tooling | 3 | partial | 4/10 | Xcommunity | |
An agent can drive the editor and project programmatically — create scenes and nodes, edit properties, trigger builds — through a documented scriptable interface Agent editor | ai-native user | Ai workflows — AI in the engine loop — agent-driven editors, copilots, codegen-friendly APIs, runtime inferenceAi workflows | 3 | partial | 4/10 | Tprobed | |
The engine ships a modern production 3D renderer — PBR materials, global illumination or baked lighting, shadows, post-processing Rendering | game developer | Core engine — the engine core — rendering, 2D, physics, performance at scaleCore engine | 3 | partial | 4/10 | Xcommunity | |
Define rules that trigger actions automatically on events G | ai-native user | Automation depth — how much of the product can run unattendedAutomation depth | 3 | partial | 3/10 | Xcommunity | |
Extend and automate the editor itself — custom tools, editor scripts, plugins that manipulate scenes and assets programmatically Extensibility | game developer | Editor tooling — the editor as a product — scene tools, extensibility, team workflowsEditor tooling | 3 | none | 0/10 | ||
Prevent my data from being used to train AI models G | ai-native user | Privacy posture — data-handling and privacy storiesPrivacy posture | 3 | n/a | untested | none yet | |
Read the product's source under an open license G | ai-native user | Openness — open source, data portability, and self-hosting storiesOpenness | 2 | full | 9/10 | Tprobed | |
The engine installs from a package registry into my existing toolchain — a library I import, bundle, and tree-shake like any dependency Web | web developer | Platform export — shipping everywhere — desktop, mobile, console, browser payloadsPlatform export | 2 | full | 9/10 | Tprobed | |
Read and modify the full engine source when I hit a wall Source | game developer | Licensing openness — the terms you build on — licenses, royalties, source access, pricing stabilityLicensing openness | 2 | full | 7/10 | Tprobed | |
Tests run headlessly — unit and integration tests of game code execute in CI against the real engine Testing | game developer | Headless automation — the engine without a human — CLI builds, CI test runs, dedicated serversHeadless automation | 2 | full | 7/10 | Cclaimed | |
The pricing and license terms have a track record of stability — no retroactive changes that reprice games already shipped Trust | studio lead | Licensing openness — the terms you build on — licenses, royalties, source access, pricing stabilityLicensing openness | 2 | full | 7/10 | Tprobed | |
Game logic runs headless on servers — a dedicated-server or server-runtime build without rendering Server | game developer | Headless automation — the engine without a human — CLI builds, CI test runs, dedicated serversHeadless automation | 2 | partial | 6/10 | Cclaimed | |
LLM-generated code mostly works on the first try — stable, well-documented APIs with abundant public examples the models have seen Codegen | ai-native user | Ai workflows — AI in the engine loop — agent-driven editors, copilots, codegen-friendly APIs, runtime inferenceAi workflows | 2 | partial | 6/10 | Tprobed | |
My game runs in the browser with a reasonable payload — WebGL today, and a credible WebGPU story Web | web developer | Platform export — shipping everywhere — desktop, mobile, console, browser payloadsPlatform export | 2 | partial | 6/10 | Xcommunity | |
Built-in physics — rigid bodies, collisions, raycasts — works out of the box Physics | game developer | Core engine — the engine core — rendering, 2D, physics, performance at scaleCore engine | 2 | disputed | 5/10 | Dcontradicted | |
Do everything through the API that I can do in the UI G | ai-native user | Openness — open source, data portability, and self-hosting storiesOpenness | 2 | partial | 5/10 | Cclaimed | |
Standard interchange formats — glTF, FBX, USD — import cleanly Formats | technical artist | Asset pipeline — getting content in — automated import, formats, marketplacesAsset pipeline | 2 | partial | 5/10 | Xcommunity | |
A large asset store or package ecosystem gives me ready-made models, tools, and plugins Marketplace | game developer | Asset pipeline — getting content in — automated import, formats, marketplacesAsset pipeline | 2 | partial | 4/10 | Tprobed | |
Asset import is automatable — import hooks, presets, and pipeline scripts process incoming assets without hand-clicking each one Import | technical artist | Asset pipeline — getting content in — automated import, formats, marketplacesAsset pipeline | 2 | partial | 4/10 | Cclaimed | |
Profile and scale — a real profiler, plus data-oriented paths (ECS, jobs, instancing) when scenes get heavy Performance | game developer | Core engine — the engine core — rendering, 2D, physics, performance at scaleCore engine | 2 | partial | 4/10 | Xcommunity | |
The edit-run loop is fast — hot reload or near-instant preview after a script change Iteration | game developer | Scripting — writing the game — languages, visual scripting, iteration speedScripting | 2 | partial | 4/10 | Cclaimed | |
2D is a first-class workflow — sprites, tilemaps, 2D physics — not a 3D afterthought 2d | game developer | Core engine — the engine core — rendering, 2D, physics, performance at scaleCore engine | 2 | partial | 3/10 | Xcommunity | |
An official AI assistant inside the engine helps with engine tasks — generating scripts, answering API questions, scaffolding scenes or assets Copilot | ai-native user | Ai workflows — AI in the engine loop — agent-driven editors, copilots, codegen-friendly APIs, runtime inferenceAi workflows | 2 | none | 0/10 | ||
Build gameplay logic with visual scripting without writing code Visual | technical artist | Scripting — writing the game — languages, visual scripting, iteration speedScripting | 2 | none | 0/10 | ||
Choose where my data is stored (region/residency) G | ai-native user | Privacy posture — data-handling and privacy storiesPrivacy posture | 2 | n/a | untested | none yet | |
Control data retention and deletion G | ai-native user | Privacy posture — data-handling and privacy storiesPrivacy posture | 2 | n/a | untested | none yet | |
Opt out of telemetry and usage tracking G | ai-native user | Privacy posture — data-handling and privacy storiesPrivacy posture | 2 | n/a | untested | none yet | |
Perform bulk operations across many items at once G | ai-native user | Automation depth — how much of the product can run unattendedAutomation depth | 2 | none | untested | none yet | |
Schedule recurring jobs or workflows G | ai-native user | Automation depth — how much of the product can run unattendedAutomation depth | 2 | n/a | untested | none yet | |
The project format and tooling play well with version control and multi-person teams — mergeable scenes, diffable text formats, or built-in collaboration Collaboration | studio lead | Editor tooling — the editor as a product — scene tools, extensibility, team workflowsEditor tooling | 2 | none | untested | none yet | |
Run ML models inside the game — an official inference runtime for on-device model execution Inference | game developer | Ai workflows — AI in the engine loop — agent-driven editors, copilots, codegen-friendly APIs, runtime inferenceAi workflows | 1 | n/a | untested | none yet | |
The project's governance and funding are transparent — a foundation, published finances, or a public roadmap I can plan against Governance | studio lead | Licensing openness — the terms you build on — licenses, royalties, source access, pricing stabilityLicensing openness | 1 | none | untested | none yet | |
Version, review, and roll back my automations G | ai-native user | Automation depth — how much of the product can run unattendedAutomation depth | 1 | none | untested | none yet |
Opportunities — the stories that would move this product's scores, from its own judged verdictsOpportunitiestop 8 of 34 stories with headroom
What would move Babylon.js’s scores — derived from its own judged verdicts, biggest headroom first. Each line quotes what the judge found missing; shipping it (or evidencing it publicly) is the fix.
Agenticness — how well agents can access and operate the productDelegate tasks to a built-in AI assistant inside the product
nonemoves Built-in AIimpact 45
The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na".
Agenticness — how well agents can access and operate the productConnect an agent via an official MCP server
nonemoves agent-readyimpact 45
Babylon.js is a rendering/game engine (not an agent), so the MCP-server axis applies as an ecosystem feature, but evidence explicitly shows no official MCP server exists — only a tiny, inactive community repo with 9 stars.
Editor tooling — the editor as a product — scene tools, extensibility, team workflowsExtend and automate the editor itself — custom tools, editor scripts, plugins that manipulate scenes and assets programmatically
nonemoves PA Scoreimpact 30
Missing: editor plugin/extension API, documented editor scripting hooks, examples of third-party editor tools or automation scripts.
Agenticness — how well agents can access and operate the productGet AI-generated insights and suggestions from my data inside the product
nonemoves Built-in AIimpact 30
The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na".
Agenticness — how well agents can access and operate the productSet up automations that run autonomously in the background
nonemoves Built-in AIimpact 30
The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na".
Agenticness — how well agents can access and operate the productOperate the product with natural-language commands
nonemoves Built-in AIimpact 30
The axis applies to this product kind (peer products hold positive or none verdicts on this story), so lack of evidence for an applicable capability is "none", never "na".
Agenticness — how well agents can access and operate the productUse an official CLI
nonemoves agent-readyimpact 30
Babylon.js is a library/engine and the story asks whether it ships an official CLI for AI-native workflows; the evidence pack covers NPM packages, docs, and runtime probes but never mentions a CLI tool of any kind, official or otherwise.
Agenticness — how well agents can access and operate the productDownload a machine-readable API spec (OpenAPI or equivalent)
nonemoves API qualityimpact 30
Babylon.js exposes only human-oriented TypeDoc API references and llms.txt files; explicit probes for OpenAPI/swagger endpoints (openapi.json, swagger.json, etc.) all returned 404, showing no machine-readable API spec is published.
Showing the top 8 of 34 — every none/partial verdict in the story verdicts table is headroom.
Think a verdict is wrong? Every verdicts-table row has a Flag link — see the methodology.
Coverage map — which docs area, API section, or community source covers which judged storiesCoverage map7 surfaces · 33 covered stories
Where the cited evidence behind each covered verdict came from — the same citations the verdicts table shows, no extra judging.
llms.txt24 stories
- Point an agent at llms.txt or agent-oriented docs
- Run the product headlessly / in CI for automation
- Drive the product through a documented public API
- Build against official SDKs
- Explore an interactive API reference with runnable examples
- Test against a sandbox environment without touching production data
- An agent can drive the editor and project programmatically — create scenes and nodes, edit properties, trigger builds — through a documented scriptable interface
- LLM-generated code mostly works on the first try — stable, well-documented APIs with abundant public examples the models have seen
- A large asset store or package ecosystem gives me ready-made models, tools, and plugins
- 2D is a first-class workflow — sprites, tilemaps, 2D physics — not a 3D afterthought
- Profile and scale — a real profiler, plus data-oriented paths (ECS, jobs, instancing) when scenes get heavy
- Built-in physics — rigid bodies, collisions, raycasts — works out of the box
- The engine ships a modern production 3D renderer — PBR materials, global illumination or baked lighting, shadows, post-processing
- A full visual editor — viewport, inspector, prefabs/scene composition — is the primary way to build levels
- Build and export the project from the command line, headless, in CI — no human clicking an editor
- Game logic runs headless on servers — a dedicated-server or server-runtime build without rendering
- Tests run headlessly — unit and integration tests of game code execute in CI against the real engine
- Do everything through the API that I can do in the UI
- Self-host the core product
- One project exports to desktop, mobile, and (directly or via partners) consoles
- My game runs in the browser with a reasonable payload — WebGL today, and a credible WebGPU story
- The engine installs from a package registry into my existing toolchain — a library I import, bundle, and tree-shake like any dependency
- The edit-run loop is fast — hot reload or near-instant preview after a script change
- The primary scripting language is productive and fully exposes the engine API, with a debugger behind it
Setup docs19 stories
- Point an agent at llms.txt or agent-oriented docs
- Run the product headlessly / in CI for automation
- Drive the product through a documented public API
- Build against official SDKs
- Test against a sandbox environment without touching production data
- Rely on versioned APIs with a documented deprecation policy
- Asset import is automatable — import hooks, presets, and pipeline scripts process incoming assets without hand-clicking each one
- A large asset store or package ecosystem gives me ready-made models, tools, and plugins
- Build and export the project from the command line, headless, in CI — no human clicking an editor
- Game logic runs headless on servers — a dedicated-server or server-runtime build without rendering
- Tests run headlessly — unit and integration tests of game code execute in CI against the real engine
- The license is permissive with no royalties or per-install fees — I keep what my game earns
- Read and modify the full engine source when I hit a wall
- The pricing and license terms have a track record of stability — no retroactive changes that reprice games already shipped
- Do everything through the API that I can do in the UI
- Export all of my data in open formats and leave
- Read the product's source under an open license
- Self-host the core product
- The engine installs from a package registry into my existing toolchain — a library I import, bundle, and tree-shake like any dependency
Hacker News15 stories
- Drive the product through a documented public API
- Build against official SDKs
- Rely on versioned APIs with a documented deprecation policy
- LLM-generated code mostly works on the first try — stable, well-documented APIs with abundant public examples the models have seen
- Standard interchange formats — glTF, FBX, USD — import cleanly
- A large asset store or package ecosystem gives me ready-made models, tools, and plugins
- Define rules that trigger actions automatically on events
- 2D is a first-class workflow — sprites, tilemaps, 2D physics — not a 3D afterthought
- Profile and scale — a real profiler, plus data-oriented paths (ECS, jobs, instancing) when scenes get heavy
- Built-in physics — rigid bodies, collisions, raycasts — works out of the box
- The engine ships a modern production 3D renderer — PBR materials, global illumination or baked lighting, shadows, post-processing
- A full visual editor — viewport, inspector, prefabs/scene composition — is the primary way to build levels
- Read and modify the full engine source when I hit a wall
- My game runs in the browser with a reasonable payload — WebGL today, and a credible WebGPU story
- The primary scripting language is productive and fully exposes the engine API, with a debugger behind it
llms.txt14 stories
- Explore an interactive API reference with runnable examples
- Test against a sandbox environment without touching production data
- An agent can drive the editor and project programmatically — create scenes and nodes, edit properties, trigger builds — through a documented scriptable interface
- Standard interchange formats — glTF, FBX, USD — import cleanly
- Asset import is automatable — import hooks, presets, and pipeline scripts process incoming assets without hand-clicking each one
- A large asset store or package ecosystem gives me ready-made models, tools, and plugins
- Profile and scale — a real profiler, plus data-oriented paths (ECS, jobs, instancing) when scenes get heavy
- Built-in physics — rigid bodies, collisions, raycasts — works out of the box
- A full visual editor — viewport, inspector, prefabs/scene composition — is the primary way to build levels
- Do everything through the API that I can do in the UI
- Export all of my data in open formats and leave
- One project exports to desktop, mobile, and (directly or via partners) consoles
- The edit-run loop is fast — hot reload or near-instant preview after a script change
- The primary scripting language is productive and fully exposes the engine API, with a debugger behind it
Typedoc docs11 stories
- Run the product headlessly / in CI for automation
- Drive the product through a documented public API
- Explore an interactive API reference with runnable examples
- Test against a sandbox environment without touching production data
- Asset import is automatable — import hooks, presets, and pipeline scripts process incoming assets without hand-clicking each one
- Build and export the project from the command line, headless, in CI — no human clicking an editor
- Game logic runs headless on servers — a dedicated-server or server-runtime build without rendering
- Tests run headlessly — unit and integration tests of game code execute in CI against the real engine
- Do everything through the API that I can do in the UI
- Self-host the core product
- The edit-run loop is fast — hot reload or near-instant preview after a script change
babylonjs.com6 stories
- An agent can drive the editor and project programmatically — create scenes and nodes, edit properties, trigger builds — through a documented scriptable interface
- LLM-generated code mostly works on the first try — stable, well-documented APIs with abundant public examples the models have seen
- A large asset store or package ecosystem gives me ready-made models, tools, and plugins
- The engine ships a modern production 3D renderer — PBR materials, global illumination or baked lighting, shadows, post-processing
- A full visual editor — viewport, inspector, prefabs/scene composition — is the primary way to build levels
- Do everything through the API that I can do in the UI
Features docs5 stories
- Standard interchange formats — glTF, FBX, USD — import cleanly
- Asset import is automatable — import hooks, presets, and pipeline scripts process incoming assets without hand-clicking each one
- A large asset store or package ecosystem gives me ready-made models, tools, and plugins
- The engine ships a modern production 3D renderer — PBR materials, global illumination or baked lighting, shadows, post-processing
- Export all of my data in open formats and leave
Probe proofs — replayable recordings from the probe harnessProbe proofs
Replayable recordings from our probe harness — see the Prove-It protocol to submit one.
$curl -s https://doc.babylonjs.com/llms.txt | head -4reproduced$ curl -s https://doc.babylonjs.com/llms.txt | head -4 # Babylon.js Documentation > Babylon.js is a powerful, beautiful, simple, and open game and rendering engine packed into a friendly JavaScript framework.
$npm view @babylonjs/core versionreproduced$ npm view @babylonjs/core version \|/9.26.1 /
Business model
Free, Apache-2.0 open source web rendering engine sponsored by Microsoft; no paid tiers.
pricing ↗Score trend
How this product’s scores have moved as evidence and verdicts are re-derived — a point per change, not per day.
Try Experimental
Run it in the microterminal →Recorded agent sessions — and a live MCP handshake where the vendor ships one.
Flag
⚑ Flag a verdictThink a verdict is wrong? Opens a prefilled GitHub issue — or use the ⚑ next to any verdict above.
For agents