agentboards.org
Compare/CodeGeeX vs IntelliCode

CodeGeeXvsIntelliCode

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

CodeGeeX
Zhipu AI · IDE extension
#128
Panel
5.8
0 spec wins
Reliability
5.2
Usefulness
5.0
Cost
7.3
Longevity
5.7

“It ships for HBuilderX and DeepIn-IDE, two editors your team has never heard of and millions of developers open every morning.”

IntelliCode
Microsoft · IDE extension
#103
Panel
5.9
0 spec wins
Reliability
6.3
Usefulness
3.2
Cost
7.7
Longevity
6.5

“It completes your arguments, which is still more than most code reviews manage.”

Spec by spec

SpecCodeGeeXIntelliCode
Architecture
CategoryIDE extensionIDE extension
Runscloudlocal
Platformsmacos, linux, windowswindows, macos, linux
Context windownot documentednot documented
Protocols
MCP clientNoNo
MCP serverNoNo
Capabilities
Runs terminal commandsNoNo
Multi-file editsNoNoIntelliCode's repeated-edit suggestions apply within a file; Microsoft points to GitHub Copilot for agentic multi-file work.
Git operationsNoNo
Browser controlNoNo
Sandboxed executionNoNo
Multi-agent orchestrationNoNo
Headless / CI modeNoNo
Models
BackboneCodeGeeX, GLMIntelliCode
Bring your own modelNoNo
Local modelsNoNo
Cost
Pricing modelfreefree
Starts at$0/mo$0/mo
Free tierYesYes
Bring your own keyNoNo
Openness
Open sourceNoNo
LicenseApache-2.0proprietary
GitHub stars8,803n/a

Which one would each critic pick

CriticCodeGeeXIntelliCodePick
El Juez——not enough reviews
El Amigo6.56.5no preference
El Crítico6.05.8CodeGeeX — Inference is cloud-only with no way to point it elsewhere, so every keystroke it sees leaves the machine and the destination is not a setting you control.
El Profesor5.85.8no preference
La Inversora6.56.5no preference
La Jefa5.57.5IntelliCode — It is already installed by default in most workloads, it runs on the developer's own machine, and it costs nothing, so there is no rollout and no security review.
El Hacker4.53.5CodeGeeX — The model repository is Apache-2.0 and the plugin I would actually run is not open at all, which is the login-shaped hole I complain about every time.

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.