Skip to content
Research · runtime comparisonupdated 2026-09-06

What each runtime can actually do.

Claude Code, Codex, and ZCode compared across 17 capabilities — compiled from each runtime's own changelogs and self-reports. Not a benchmark: a record of what the currently profiled runtimes expose today.

claude codev2.1.263
claude-opus-5 · ~1M context · checked 2026-09-06
codexCLI 0.153.4
GPT-6 Astra · 272K catalog window · checked 2026-09-04
zcode3.9.2
GLM-5.3 · 1M declared · checked 2026-08-26
01Core capability matrix17 capabilities · 03 of 03 runtimes

Seventeen capabilities, three columns.

Status at a glance; hover a status for the record's own words. Each runtime updates only its own column.

comparing
hover or tap a status for the record's words
Orchestration
Spawn background tasks
Spawn subagents
Parallel agent teams
Web & tools
MCP tools
Web search
URL fetch (no browser)
Browser interaction
Workspace
File system access
Multi-folder projects
Git operations
GitHub CLI (gh)
Sessions
Session memory
Interactive mode
Plan mode
Limits & cost
Context window
Cost model
Sandbox restrictions

Compiled from each runtime's self-report and current runtime changelogs. Claude checked against Claude Code v2.1.263 with claude-opus-5 (2026-09-06); Codex against stable CLI 0.153.4, GPT-6 Astra (2026-09-04); ZCode 3.9.2 as a live in-session self-report (2026-08-26, re-verified same day).

02Strengths03 runtimes

Where each one is strongest.

Best-at, ideal task type, and cost profile — the summary the dispatch decision actually uses.

claudeclaude code cli
best at

Orchestration, multi-agent teams, persistent memory, plan-then-execute.

ideal task type

Team coordination, complex multi-file refactors, long-running sessions.

cost profile

Flexible (haiku subagents for cheap tasks, opus for complex, Fable 5.1 at 2× Opus API rates for the hardest work).

codexdesktop app
best at

Terminal execution, Astra workflows across code/apps/research, GPT-5.6 task tiers, iterative patching, native delegation and task coordination, browser verification, artifact review, and app-server automation.

ideal task type

Complex workflows spanning code, browser, and documents; shell-heavy implementation, targeted edits, PR follow-up, artifact review, cloud follow-up, and app-server prototypes; asynchronous clarification can keep independent work moving.

cost profile

Astra targets the hardest workflows; Sol, Terra, and Luna provide other quality/cost choices. ChatGPT credits, conditional task estimates, and published API token rates measure different things.

zcodedesktop app
best at

Agentic coding with GLM-5.3, plugin-driven browser and desktop automation, office document production, scheduled automations.

ideal task type

Full-stack coding with GUI verification, document generation pipelines (PDF/Office), cross-session handoff workflows, MCP integrations.

cost profile

Token-based; details not surfaced in-session (unverified).

03Parity gaps05 gaps · actionable

Where one runtime's limit blocks the loop.

The highest-impact gaps, each with its proposed fix. A gap without a fix is a complaint.

No shared team/task-list primitive

codex · zcode

Codex adds a task dashboard, queue, mentions, and parallel delegation but still lacks Claude-style named teammates and shared task lists (the latter opt-in on current Claude models); ZCode has concurrent subagents and agent-to-agent messaging but no shared team primitive.

fix → Agent cards — let runtimes discover and delegate to capable peers.

Memory asymmetry

codex (partial) · zcode (handoff only)

Cross-session context degrades without a MEMORY.md equivalent.

fix → Standardize memory protocol; each runtime implements its own persistence layer.

Sandbox blocks git/gh

claude

Every git/gh call needs sandbox configuration.

fix → Configure sandbox allowlists (excludedCommands globs such as git *, gh *) or disable sandbox for specific commands.

Full browser automation gap

claude without the Chrome extension · codex without a configured surface

Claude falls back to read-only WebFetch when the extension is absent or the session is API-key-authenticated; Codex browser control depends on an enabled surface and an approved context.

fix → Install the Claude in Chrome extension (/chrome), or delegate to a browser-capable runtime (configured Codex, or ZCode) after privacy review.

Public skill coverage

zcode

kiloloop/oacp-skills ships claude/ and codex/ variants for all 5 public skills; no zcode/ variants — ZCode users must rely on convention-based adoption.

fix → Add zcode/ variants to each public skill, or document the convention-based pattern as a first-class install path.

04The full record08 sections · 01 file

The page is the digest. The file is the record.

Five more sections live in the repository — versioned, diffable, and updated by the changelog skills, not by hand.

§2Distinctive capabilities24 rows
§3Public OACP skills coverage5 skills × 3 runtimes
§5Known limitations7 dimensions
§7Additional dimensions16 rows
§8Source notesfull provenance

Each runtime should update only its own column. Discrepancies should be resolved by the runtime owner.