Agentic Index
Google Jules vs OpenAI Codex (2026)
Google Jules last verified ·OpenAI Codex last verified
Google Jules is a cloud only agent for individual developers, while OpenAI Codex runs on your own machines as well as in the cloud, with controls for a whole company. That verdict is the Agentic Index coverage score, graded from each vendor's own published materials.
Each takes a coding task into a cloud sandbox and hands back a pull request. Jules clones the repository into a Google Cloud virtual machine and shows a plan you can edit before it writes code. It then runs the tests, fixes failing CI checks and opens the pull request, working on Gemini models. Jules is free for 15 tasks a day, or 100 a day with Google AI Pro at $19.99 a month. Codex works in the ChatGPT desktop app, a CLI, an IDE extension, the web and the cloud, under a sandbox and approval policy you set. It adds skills, plugins, memories and computer use, and it is included in every ChatGPT plan, from Free and Go at $8 through Plus at $20 and Business at $20 a user. Choose Google Jules when one developer wants a free cloud agent that shows its plan first. Choose OpenAI Codex when a team needs the agent on its own machines under company controls.
On the Agentic Index coding agent ranking, neither Google Jules nor OpenAI Codex clears the bar, which asks for all five merge loop capabilities documented in full. Neither documents knowledge grounding and RAG in full, nor testing, debugging and optimization. 3 of the 96 vendors in the lane clear it. See the coding agent ranking
This comparison is published by Agentic Index, an independent agentic AI vendor research platform. Google Jules and OpenAI Codex are each graded against the same 14 capability Agentic Index taxonomy, from the vendor's own public materials under the Agentic Index verification standard, alongside 1,165 researched vendors. No vendor pays for placement and no vendor has reviewed this page. How this evidence is graded
This comparison, as a video
Google Jules vs OpenAI Codex in 5 minutes
A short walkthrough of what each product does, how they are priced, where they overlap and when to choose each.
Choose Google Jules if
- You want to read and edit the agent's plan before it writes any code, with a critic reviewing each change.
- Fifteen free cloud tasks a day, or 100 a day with Google AI Pro at $19.99 a month, cover how much you delegate.
- You want failing GitHub Actions checks on the agent's pull requests fixed and resubmitted without asking.
Choose OpenAI Codex if
- You want the agent on developers' own machines as well as in the cloud, under a sandbox and approval policy you control.
- You need company controls such as SCIM, managed configuration, a Compliance API and OpenTelemetry export, backed by OpenAI's SOC 2 and ISO reports.
- Your team already pays for ChatGPT, so Codex comes with the plan you have.
| Feature | G Google Jules |
O OpenAI Codex |
|---|---|---|
| Action & orchestration | ||
|
Integrations & Tool Calling Ability to connect agents to real systems through native integrations, OAuth-authenticated actions, custom tools, APIs, webhooks, or MCP-compatible tools. |
||
|
Google JulesIntegrations & Tool Calling Deep GitHub integration covers issue labels, branches, pull requests and reading and responding to PR comments, MCP servers connect Linear, Stitch, Neon, Tinybird, Context7 and Supabase by API key in Settings, a Render integration detects failed builds and pushes fixes, and the REST API wires Jules into Slack, Linear, Jira and CI pipelines including GitHub Actions. GitHub is the only source control host. The agent runs shell commands, installs dependencies and executes builds inside its VM with repository level environment variables. Sourcejules.google/docs/changelog 2026-02-02 and 2025-10-03, docs/integrations/renderread 2026-10-28 |
||
|
OpenAI CodexIntegrations & Tool Calling MCP connects outside services, with approved server lists under enterprise control, and plugins and connectors extend the agent with tool calls that have side effects, which require approval when they advertise destructive annotations. Third party integrations cover GitHub, GitLab, Slack and Linear. The agent runs shell commands under an OS sandbox, uses a web search tool with cached, live and disabled modes, and reaches the browser and computer use surfaces. Sourcelearn.chatgpt.com/codex/extend/mcp, /codex/third-party and /codex/agent-approvals-securityread 2026-08-30 |
||
|
Workflow Orchestration Ability to sequence, branch, retry, route, and combine deterministic workflow nodes with autonomous agent steps. |
||
|
Google JulesWorkflow Orchestration A task runs as a multi step plan the agent produces before writing code, with an Interactive Plan mode where it asks clarifying questions and refines the approach collaboratively, a Critic agent performing adversarial review of every proposed change during generation, a Planning Critic reviewing auto approved plans and reported to cut task failure rates by 9.5 percent, a CI Fixer loop that detects failures, fixes, commits and resubmits, and up to five parallel sessions on one prompt via the CLI --parallel flag, within the plan's concurrency limit of 3, 15 or 60 tasks. Sourcejules.google/docs/changelog 2026-01-26-1, 2026-02-19 and 2025-11-10, docs/cli/reference and docs/usage-limitsread 2026-10-28 |
||
|
OpenAI CodexWorkflow Orchestration The agent sustains long autonomous sessions across many sequential tool calls, with configurable subagents, git worktrees for parallel work, projects and chats organizing multi step work, long running work and scheduled tasks, and the desktop app running agent threads side by side. The cloud agent works asynchronously in isolated containers so many tasks proceed in parallel, and multi agent tooling is among the stable CLI surfaces. Sourcelearn.chatgpt.com/codex/agent-configuration/subagents, /codex/environments/git-worktrees and /codex/long-running-workread 2026-08-30 |
||
|
Triggers & Channel Coverage How agents wake up and where they work: schedules, webhooks, message events, CRM events, inbox events, chat, email, voice, and collaboration tools. |
||
|
Google JulesTriggers & Channel Coverage Tasks start from a jules label on a GitHub issue, the web app, the Jules Tools CLI, a REST API for CI/CD and chat integrations, scheduled recurring tasks (daily, weekly or monthly) with pause and resume, proactive suggested tasks where Jules scans a repository for TODO comments and surfaces work, pull request comments which the agent reads and acts on, and automatic triggers on CI failures in GitHub Actions and failed Render builds. No inbound webhook is documented. Sourcejules.google/docs/guides/scheduled-tasks, docs/guides/running-tasks and docs/changelog 2025-12-10, 2026-01-26 and 2026-02-19; jules.google/docs/changelog/2025-12-101read 2026-10-28 |
||
|
OpenAI CodexTriggers & Channel Coverage Codex runs in the ChatGPT desktop app, a Codex CLI, an IDE extension, ChatGPT on the web, Codex cloud, and a Remote mode driven from a phone. Scheduled tasks run work on a timetable, long running work is supported, a GitHub Action and non-interactive mode cover CI, and third party integrations reach GitHub, GitLab in beta, Slack and Linear, with notifications closing the loop. Sourcelearn.chatgpt.com/codex/automations, /codex/github-action and /codex/third-partyread 2026-08-30 |
||
| Knowledge & context | ||
|
Knowledge Grounding & RAG Ability to ground agent behavior in company data through document ingestion, retrieval, external knowledge APIs, semantic search, or RAG layers. |
||
|
Google JulesKnowledge Grounding & RAG Each task clones the repository into its VM and reads it directly, with AGENTS.md conventions read from the repository root, a file selector to scope context, image upload and web search for current documentation. Suggested Tasks scan up to five repositories for TODO comments. Context is assembled per task, and the docs describe no persistent index, graph or embeddings layer over the customer's code. Sourcejules.google/docs, docs/guides/suggested-tasks and docs/environmentread 2026-10-28 |
||
|
OpenAI CodexKnowledge Grounding & RAG Context is assembled for each run. The agent reads the repository, resolves @ mentioned files, calls MCP servers for outside sources and uses a web search tool whose cached mode draws on a web index OpenAI maintains. AGENTS.md and Rules give the agent standing instructions. OpenAI does not describe an index that Codex keeps over the customer's own code or documents. Sourcelearn.chatgpt.com/docs/agent-configuration/agents-md, /docs/extend/mcp and the docs llms.txt indexread 2026-09-29 |
||
|
Memory & State Persistence Ability to persist context across a run, conversation, workflow, user, team, or longer-term memory layer. |
||
|
Google JulesMemory & State Persistence Repository Memory saves the user's preferences, nudges and corrections during a task and references them on later tasks in that repository, toggled on or off per repository under Knowledge. The changelog is the only page that describes it. Memory has no stated lifetime and no documented way to review or delete entries, and environment snapshots and variables are setup state, not memory. Sourcejules.google/docs/changelog 2025-09-30read 2026-10-28 |
||
|
OpenAI CodexMemory & State Persistence Codex writes its own memories: once enabled, it turns useful context from eligible prior chats into local memory files under CODEX_HOME, redacting secrets and updating in the background after a chat goes idle. Memories live in the local Codex home, /memories chooses per chat whether a chat may read existing memories or feed future ones, and managed configuration can pin the feature on or off. Unused memories retire from consolidation after memories.max_unused_days, 30 by default and capped at 365, memories.max_rollout_age_days bounds the chats considered, and the files can be opened for review. The feature is off by default. AGENTS.md holds standing rules, separate from memory. Sourcelearn.chatgpt.com/docs/customization/memories and /docs/config-file/config-referenceread 2026-09-29 |
||
| Control & trust | ||
|
Human Oversight & Guardrails Approval steps, consent checkpoints, escalation rules, structured guardrails, policy constraints, and pause/resume controls. |
||
|
Google JulesHuman Oversight & Guardrails The agent writes a step by step plan the user approves, edits or rejects before any code is written, with an Interactive Plan mode asking clarifying questions first, and the user can chat with it at any point. If the user leaves, the plan auto approves on a timer, and the API's requirePlanApproval option is off by default, so a Planning Critic reviews plans that are auto approved. Work lands on a branch and the user chooses when to publish or open a pull request, a Critic agent adversarially reviews each proposed change during generation, tasks can be paused, resumed or deleted mid run, and commit authorship is configurable so changes are attributed to Jules, co-authored, or the user. Sourcejules.google/docs/guides/review-plan, docs/api/reference and docs/changelog 2026-01-26-1 and 2026-02-19; jules.google/docs/changelog/2025-08-082read 2026-10-28 |
||
|
OpenAI CodexHuman Oversight & Guardrails Two independent layers govern the agent: sandbox mode controlling what it can technically do, and approval policy controlling when it must ask, with presets from read-only through workspace-write to danger-full-access and a granular policy that keeps chosen approval categories interactive while auto rejecting others. An Auto-review reviewer agent can evaluate eligible approval requests against a published policy checking for data exfiltration, credential probing, security weakening and destructive actions, failing closed on parse or session errors. Destructive MCP and app tool calls always require approval, .git, .agents and .codex are protected read-only inside writable roots, and enterprise managed requirements override local settings. Sourcelearn.chatgpt.com/codex/agent-approvals-security and /codex/sandboxing/auto-reviewread 2026-08-30 |
||
|
Security, Identity & Governance RBAC, SSO, auditability, encryption, least-privilege tool access, compliance posture, and data handling policy. |
||
|
Google JulesSecurity, Identity & Governance Jules isolates each task in a cloud VM, reaches repositories only as granted through the GitHub app, stores integration API keys encrypted, does not train on private repositories and tells users they are responsible for the code they run. There is no access surface (SSO, roles, organization administration) and no attestation or trust page, and paid plans are sold to individual Google accounts only. Sourcejules.google/docs/faq and docs/usage-limitsread 2026-10-28 |
||
|
OpenAI CodexSecurity, Identity & Governance OpenAI's trust portal at trust.openai.com lists SOC 2 Type 2, SOC 3, ISO/IEC 27001, 27017, 27018, 27701 and 42001, CSA STAR and FedRAMP 20x, with the SOC 2 report scoped to the API Platform, ChatGPT Enterprise, ChatGPT Edu and ChatGPT Team, the plans and API key path Codex runs on, though the portal does not name Codex separately. Access controls include managed configuration that overrides local settings, roles and workspace permissions, SCIM and EKM on Enterprise, workload identity federation, service accounts, plugin, connector and skill controls, approved MCP server lists, a Compliance API with audit events, Private Link, IP allowlisting and mutual TLS. Sourcetrust.openai.com and learn.chatgpt.com/docs/enterprise/managed-configurationread 2026-09-29 |
||
|
Observability & Auditability Traces, logs, execution histories, metrics, audit events, and debugging detail for production agent behavior. |
||
|
Google JulesObservability & Auditability Each session carries a collection of activities representing the steps the agent took. Google calls them immutable, following an event source pattern and usable to reconstruct the session, and they are retrievable through the REST API with a createTime range cursor. The web app shows a real time activity feed, per task logs and task history per repository, the full changeset is downloadable in git patch format, and the critic agent's step by step evaluation is visible in the UI as it runs. No organization audit log is documented. Sourcejules.google/docs/changelog 2026-01-26-4 and 2025-09-03, docs/guides/coderead 2026-10-28 |
||
|
OpenAI CodexObservability & Auditability Opt in OpenTelemetry export emits an event catalog that includes codex.tool_decision, recording approved or denied and whether the source was configuration or the user, codex.tool_result with duration, success and output snippet, codex.api_request, codex.sse_event and codex.user_prompt, with counter and duration histogram metrics, exportable over OTLP to a collector the customer controls. Enterprise adds a Compliance API and audit events, workspace analytics and an Analytics API. Telemetry is off by default and prompt text is redacted unless explicitly enabled. Sourcelearn.chatgpt.com/codex/agent-approvals-security and /codex/enterprise/compliance-apiread 2026-08-30 |
||
|
Deployment & Data Residency Deployment modes and options, including SaaS, dedicated cloud, VPC, on-prem, hybrid, local runtime, and self-hosting. |
||
|
Google JulesDeployment & Data Residency Every task runs in a fresh, short lived Ubuntu VM in Google's cloud with internet access, whose toolchain the customer shapes with a setup script, snapshots and environment variables. That configures the environment, not where it runs. There is no region choice, customer VPC, on premises or tenant option, so a customer with a different residency requirement has no other answer. Sourcejules.google/docs/environment and docs/faqread 2026-10-28 |
||
|
OpenAI CodexDeployment & Data Residency Local environments run entirely on the developer's machine through the CLI, IDE extension and desktop app under an OS enforced sandbox using Seatbelt on macOS, bwrap and seccomp on Linux and a native Windows sandbox or WSL2, with dev container support and a published secure devcontainer reference; cloud environments run in isolated OpenAI managed containers with configurable internet access and domain allow lists. Enterprise adds Private Link, IP allowlisting, mutual TLS, IP egress ranges and Amazon Bedrock as a model provider. Sourcelearn.chatgpt.com/codex/environments/modes, /codex/agent-approvals-security and /codex/amazon-bedrockread 2026-08-30 |
||
| Solution readiness | ||
|
Prebuilt Agents, Templates & Packs Ready-made workflows, packaged employees, templates, blueprints, industry solutions, and role-specific agents that reduce time-to-value. |
||
|
Google JulesPrebuilt Agents, Templates & Packs Sample prompts ship as one click starting points, three scheduled task templates (Performance, Design and Security) start recurring tasks, suggested tasks let the agent propose work it has found in the repository, and MCP integrations are offered from a vendor curated list of vetted servers. There is no library of prebuilt agents, installable skills or marketplace, and the community prompt list lives in a GitHub repository. Sourcejules.google/docs/guides/scheduled-tasks, docs and docs/changelog 2025-09-02 and 2026-02-02; jules.google/docs/changelog/2025-09-02read 2026-10-28 |
||
|
OpenAI CodexPrebuilt Agents, Templates & Packs Skills and plugins extend Codex, with dedicated build guides, a plugin architecture and submission path, enterprise plugin management, and plugin and skill controls for administering approved catalogs, and a Claude Code plugin can be submitted as well. Subagents are configurable per project, and Codex Micro ships as a packaged capability. Sourcelearn.chatgpt.com/codex/skills-and-plugins, /codex/build-plugins and /codex/enterprise/plugin-managementread 2026-08-30 |
||
| Platform extensibility | ||
|
Model Flexibility & Routing Ability to work across multiple foundation models, route tasks to different models, or let buyers bring their own providers and keys. |
||
|
Google JulesModel Flexibility & Routing Every model Jules runs is Google's Gemini. The changelog makes Gemini 3 Flash the base model on all tiers and Gemini 3.1 Pro the default for Pro subscribers, which is a tier choice within one maker, while the homepage and usage page still name Gemini 2.5 Pro and Gemini 3 Pro. There is no user model choice, no other maker and no bring your own model path. Sourcejules.google/docs/changelog 2026-03-09 and 2026-01-30, docs/usage-limitsread 2026-10-28 |
||
|
OpenAI CodexModel Flexibility & Routing Local Codex clients let the customer choose the model maker, accepting custom model providers in config.toml, each defined by base URL, wire API and authentication, and OpenAI's own example configures Mistral's API and a local Ollama endpoint beside an OpenAI proxy. OSS mode runs Codex against a local open source provider, Ollama or LM Studio, chosen per run with --local-provider or set as the default. Within OpenAI's own family the model is selectable per session and per workspace. Amazon Bedrock is a built in provider that serves OpenAI models on AWS. Sourcelearn.chatgpt.com/docs/config-file/config-advanced (Custom model providers, OSS mode) and /docs/modelsread 2026-09-29 |
||
|
APIs, SDKs & MCP Extensibility Composability layer: stable APIs, SDKs, MCP tool consumption/serving, custom tools, and integration into internal systems. |
||
|
Google JulesAPIs, SDKs & MCP Extensibility A REST API at jules.googleapis.com/v1alpha, in alpha, authenticates with an X-Goog-Api-Key header (up to three keys) and exposes Sources, Sessions (create, list, approve plan, send message, auto create PR, repoless sessions) and Activities with git patch output. A Jules Tools CLI installs via npm as @google/jules with jules remote new, list and pull, parallel execution and local patch application, a Gemini CLI extension exposes Jules as a slash command, and MCP server support connects a vetted set of servers by API key. Sourcejules.google/docs/api/reference, docs/api/reference/authentication, docs/cli/reference and docs/changelog 2026-02-02; jules.google/docs/changelog/2026-02-02read 2026-10-28 |
||
|
OpenAI CodexAPIs, SDKs & MCP Extensibility Developers build on a Codex SDK, an App Server, an MCP Server, a GitHub Action and a non-interactive mode, alongside MCP client support with approved server lists, hooks that fire on lifecycle events, Site tools via WebMCP, Record and Replay, and a Python SDK with auth. The CLI is open source in Rust at github.com/openai/codex with a published default reviewer policy, and the documentation offers an llms.txt index and a markdown version of every page. Sourcelearn.chatgpt.com/codex/codex-sdk, /codex/mcp-server and /codex/open-sourceread 2026-08-30 |
||
|
Testing, Debugging & Optimization Testing, debugging, scoring, retries, fallbacks, quality gates, and optimization loops for improving agent workflows before and after deployment. |
||
|
Google JulesTesting, Debugging & Optimization The agent runs the project's existing test suite and iterates when tests fail, writes tests on its own, and the CI Fixer detects a failed GitHub Actions check, works through a fix and resubmits the pull request in a loop. A Critic agent adversarially reviews proposed changes and a Planning Critic reviews plans, both internal quality mechanisms, and Run to Validate checks environment setup scripts. These verify the customer's code and the agent's own output. There is no harness for customers to evaluate or regression test agent behavior. Sourcejules.google/docs/changelog 2026-02-19 and 2026-01-26-1, docs/environmentread 2026-10-28 |
||
|
OpenAI CodexTesting, Debugging & Optimization Codex reviews code changes before commit, a Codex Security plugin and CLI run scans, deep scans and bulk scans in CI with a security workbench and finding export, and the agent runs tests and iterates on failures; Auto-review evaluates individual agent actions against a risk policy at runtime. Those tools check the customer's code and gate the agent's actions, and there is no harness to evaluate, replay or regression test the agent's own behavior. Sourcelearn.chatgpt.com/codex/code-review and /codex/securityread 2026-08-30 |
||
| Specialist automation | ||
|
Browser & Computer Use Browser, desktop, or remote/local computer control for workflows that cannot be handled through stable APIs alone. |
||
|
Google JulesBrowser & Computer Use The default Jules base image includes Playwright and ChromeDriver for front end testing, and the agent renders a running web application and returns a screenshot of the front end alongside its code changes when asked to verify its work, with images shown in the diff viewer. With Gemini 3 Pro, visual verification of web app outcomes is more precise, and Jules can also search the web for documentation and code examples during a task. Sourcejules.google/docs/changelog 2025-08-07 and 2025-11-19, docs/environmentread 2026-10-28 |
||
|
OpenAI CodexBrowser & Computer Use Browser and Computer Use let the agent work in a browser and operate a computer, alongside a browser extension for signed in browser workflows and Appshots. Browser and Computer Use activity is treated as its own traffic surface outside the command network proxy, with its own feature settings and workspace policies, and managed configuration lets administrators set feature requirements for browsers and Computer Use. Sourcelearn.chatgpt.com/codex/computer-use, /codex/browser and /codex/agent-approvals-securityread 2026-08-30 |
||
Pricing snapshot
Sourced from the Index pricing dataset · open each vendor's profile for full detail.
| Pricing | G Google Jules |
O OpenAI Codex |
|---|---|---|
|
Entry price Lowest public entry point |
Free · Jules in Pro via Google AI Pro $19.99/mo | Included in ChatGPT plans. Free is $0 and Go is $8 a month, with GPT-6 Luna in the desktop app subject to rollout. Plus is $20, Pro starts at $100 ($100, $200 or $500), and Business is $20 per user a month billed annually ($25 monthly). Enterprise and Edu are sold through sales, and an API key pays per token. |
|
Pricing confidence How public the numbers are |
Public, exact | Public, partial |
|
Billing Primary billing axis |
hybrid | A ChatGPT plan subscription with an included Codex usage quota, plus usage beyond the quota paid with ChatGPT credits or, on the API key path, at per token API rates. |
|
Variable cost Workload / overage exposure |
Low variable cost | High variable cost |
|
Free tier / trial Try before you buy |
Free tier
|
Free tier
|
|
Buying motion Self-serve vs sales call |
Self-serve | Self-serve |
|
Pricing verified When this pricing was last checked |
Product updates
Newest first, each linked to its source.
Google Jules
-
· Agent capability
Sessions in Jules can continue until November 11, 2026, after which Jules turns read only, and the product winds down for good on December 31, 2026 as Google unifies its agentic coding work.
Source
OpenAI Codex
-
· MCP / tool calling / API
Codex CLI 0.164.0 lets enterprise MCP connections authenticate with eligible Codex account grants or a dedicated sign in, including OAuth clients that require secrets.
Source -
· Deployment / data residency
Codex CLI 0.161.0 lets the Bedrock Mantle provider accept AWS GovCloud regions, so Codex can send inference to OpenAI models hosted in Bedrock GovCloud.
Source -
· Agent capability
OpenAI shipped GPT-6.1 Sol into Codex on desktop and CLI, describing it as near top tier performance at lower cost for long running work, and made it the default model in the CLI.
Source
Jules's paid tiers come with Google AI Pro and Ultra subscriptions and are open only to personal @gmail.com accounts, with Workspace accounts not yet supported.
More comparisons with Google Jules or OpenAI Codex
Other matchups in coding agents
Not the pairing you were after? These compare a different set of coding agents on the same 14 capabilities.