Sim vs Vellum
Sim is the open-source AI workspace where teams build, deploy, and manage AI agents visually, conversationally, or with code. Here is how Sim compares to Vellum on platform architecture, AI capabilities, integrations, pricing, security, and support. Every fact below is sourced and dated.
Sim is an open-source AI workspace for building, deploying, and managing AI agents. This page compares Sim to Vellum across platform architecture, AI capabilities, integrations, pricing, security and compliance, observability, and support, using sourced, dated facts for buyers evaluating both platforms.
What is Sim?
Sim is the open-source AI workspace where teams build, deploy, and manage AI agents, connecting 1,000+ integrations and every major LLM to automate real work visually, conversationally, or with code.
What is Vellum?
Vellum is an enterprise AI development platform for building, evaluating, and deploying LLM prompts, workflows, and agents.
Sim vs Vellum: feature-by-feature comparison
CompareSim vs Vellum
Sim
Vellum
Platform
Builder type
Sim
Visual canvas, chat, or codeVisual drag-and-drop canvas, natural-language (Chat), or code (API/SDK)
Vellum
Visual builder, code SDK, or natural languageVisual workflow builder plus a code-first SDK, and a natural-language 'Vellum for Agents' mode for non-technical users. Vellum offers a visual graph builder and a code-first SDK that stay in sync, with nodes for model calls, retrieval, tool/API steps, and control flow. A separate 'Vellum for Agents' surface lets non-engineers describe a goal in plain language to generate an agent.
Learning curve
Sim
Low, plus natural-language Chat for non-technical usersLow for visual building; natural-language Chat surface (with a workflow-scoped Copilot) for non-technical builders. Chat lets users describe a workflow in plain language and have Sim scaffold or edit it; the visual canvas requires no code to connect blocks.
Vellum
Not publicly documentedUnknown. Not publicly documented.
Self-hosting
Deployment options
Sim
Cloud-hosted or self-hosted, no mid-tier VPC optionCloud-hosted (managed, multi-tenant SaaS) or self-hosted (Docker/Kubernetes). No documented managed single-tenant/VPC hosting tier in between. The Enterprise plan's only hosting-related row in the pricing comparison table is a boolean "Self Hosting" flag; there is no dedicated-instance/VPC offering.
Vellum
Managed, self-hosted, or VPC installVellum Managed (fully managed hosted service), self-hosted, and Vellum VPC install on AWS, Azure, or GCP
Templates
Sim
Vellum
Exists; contents undocumentedA templates page exists on vellum.ai; contents are not publicly detailed.
License
Sim
Apache 2.0Apache License 2.0
Vellum
Proprietary enterprise platform; MIT consumer appProprietary/commercial for the core enterprise Vellum platform. A separate consumer 'Vellum Assistant' product is open-sourced under MIT license.. github.com/vellum-ai/vellum-assistant is MIT-licensed (confirmed live, 825+ stars). No evidence the enterprise workflow/evaluations platform itself is open source.
Environment promotion
Sim
Yes: fork a whole workspace into a dev/qa/prod-style child, diff it, and promote or roll back changes in either direction. Credential and env-var remapping is required before every promote, so secrets are never silently copied across environments. Gated to Enterprise plan on hosted Sim, or a FORKING_ENABLED flag on self-hosted deployments.
Vellum
Dev/Staging/Prod with one-click promotionVellum has Development/Staging/Production environments, each with isolated API keys, release histories, and environment variables/secrets. A 'Promote' button moves a tested Release to another environment, or you can deploy directly to multiple environments at once. Sandboxes (prompt/workflow definitions) are shared across environments; deployments and releases are environment-scoped.. Promotion operates at the level of individual workflow/prompt releases rather than whole-workspace forking.
Version control
Sim
Deployment rollback plus Copilot edit diff/revertDeployed-version history with rollback for every workflow; server-persisted checkpoint/revert and visual diff (accept/reject) specifically for Copilot AI edits. Manual drag-and-drop undo/redo is client-side/localStorage only (capped at 100 ops, 5 stacks), not server-synced across devices. Deployment history does not include an arbitrary version-to-version diff tool, and knowledge base documents have no version history.
Vellum
Release history with one-click rollbackVellum maintains per-environment Release history with instant one-click rollback (no code changes required) and semantic-versioned Release Tags or a rolling LATEST tag. Version comparison is mentioned, but a dedicated diff/side-by-side view is not documented.
Realtime collaboration
Sim
Vellum
No: Vellum's collaboration model is asynchronous, not live multi-user editing. Prompt Sandbox lets team members see each other's history, tag entries, and share/invite others, but there is no documented live-cursor or simultaneous editing of the same workflow or prompt.. Vellum documents shared visibility into sandbox history and sharing/invite mechanisms, framed around sequential iteration rather than simultaneous editing.
Native file storage
Sim
Yes: a native Files area with folder hierarchy, link-based sharing (public, password, email OTP, or SSO auth), and a workspace-level Recently Deleted view covering workflows, tables, knowledge bases, files, and folders. Admins can restrict which share-auth modes (public/password/email/SSO) a permission group is allowed to use.
Vellum
No: Vellum's document handling is scoped to Document Indexes for RAG (upload, indexing, search), not general-purpose file storage. No folder hierarchy, shareable links with password/SSO auth, or trash/recovery feature is documented.. Documents are environment-scoped, uploaded for indexing with size limits (up to 32MB) and supported formats (PDF, DOCX, CSV, etc.), but folder hierarchy, link-sharing with auth, and deleted-item recovery are not documented.
Native data tables
Sim
Yes: a native spreadsheet-like Tables feature (typed columns, not an external DB connector) with full keyboard support (arrow keys, Tab, copy-paste bulk load, Cmd/Ctrl+Z undo) and atomic per-row writes from multiple workflows at once. No public fixed row-limit figure is documented (guidance says paginate reads past ~100k rows); a workflow can also be wired to run per row via a "workflow column."
Vellum
No: Vellum has no native, spreadsheet-like data-table feature (persistent rows/columns, keyboard navigation, per-row writes from workflows) comparable to a built-in database. Its only tabular-data handling is uploading CSV/XLS files and extracting or generating structured output within workflow nodes.. Vellum supports CSV/XLS upload and structured JSON extraction, and has blog content on converting PDFs to CSV, but no editable in-platform spreadsheet/data-table object exists; Executions tables are read-only run logs, not a general-purpose data store.
Rich-text document editor
Sim
Vellum
Unknown for the workflow platformUnknown: no documentation describes an inline rich-text/WYSIWYG markdown editor for documents stored within the B2B Vellum workflow platform (docs.vellum.ai). Prompt/node inputs in Workflows appear to be plain text/code fields.. A rich-text/Markdown document editor is documented for the separate Vellum personal-assistant product (vellum.ai, a 2026 pivot product), but not confirmed for the B2B workflow/agent platform compared here.
Sub-workflows (composition)
Sim
Vellum
Yes: Vellum has a Subworkflow node that executes a deployed or inline workflow as a step inside a parent workflow, waiting for it to finish and passing/receiving data through defined inputs and outputs.. Both "Deployed Subworkflows" (calling a separately versioned, released workflow) and "Inline Subworkflows" (defined within the parent workflow for modularization/reuse) are documented; the Agent Node can also register subworkflows as callable tools.
Custom blocks (org-wide reuse)
Sim
Yes: a user with admin access on a workflow's workspace can publish its deployed version as a named, iconed custom block that appears in the block toolbar for the whole organization. Inputs are read live from the source workflow's Start block; the publisher picks which outputs to expose. The block always runs the source workflow's latest deployed version, and its internal steps and intermediate values are never exposed to consumers. Enterprise-gated on Sim Cloud; self-hosted deployments can opt in via a feature flag. Distinct from the Workflow (sub-workflow) block, which composes a workflow the same author controls, and from custom code/tool blocks, which run inline code rather than a full deployed workflow.
Vellum
Partial: Deployed Subworkflow node, not a published named/iconed toolbar blockPartial: Vellum has no documented feature for publishing a workflow as a named, iconed block that appears in a shared toolbar for the whole organization. Its closest mechanism is the Deployed Subworkflow node, a dedicated 'Deployed Subworkflow' option in the Insert Node Panel that lets a builder insert an already-deployed workflow as a single node in their own separate workflow, wiring only its defined inputs/outputs rather than exposing the source workflow's internal nodes. Pointing it at the LATEST release tag means it automatically runs the source workflow's newest deployment with no separate republish step.. Unlike Sim's Custom Blocks, there is no documented step where a publisher assigns a custom name/icon that then surfaces the workflow as a first-class item in the general node toolbar alongside built-in nodes; consumers instead search for and link a specific deployed workflow through the Subworkflow node type. There is also no documented ability to hand-pick and rename a subset of outputs to expose, and no Enterprise-only gating or block-level allow/deny governance specific to this reuse mechanism beyond Vellum's general workspace-wide RBAC roles.
Pricing
Pricing model
Sim
Credit-based billing, BYOK exempt from capsCredit-based usage billing (Stripe), with bring-your-own-key exemption from metered caps
Vellum
Not publicly listed; third-party analysis describes credit-based tiersNot publicly listed for the B2B enterprise platform: Vellum's current pricing page has been fully replaced by the unrelated consumer 'Personal Intelligence' product's plans. A September 2025 third-party pricing analysis, published before that switch, described the enterprise platform as execution-credit-based with per-tier seat caps (Free, Pro, Enterprise) rather than the prepaid-credit model now shown.. Current pricing pages (vellum.ai/pricing, vellum.ai/docs/pricing) describe the consumer 'Personal Intelligence' product's plans (Base/Free and Pro $50/mo tiers with configurable vCPU/RAM/storage add-ons), not the enterprise workflow platform. The last third-party description of the enterprise platform's model itself states Vellum 'does not publicly list pricing details on its website,' so the figures below are third-party estimates, not Vellum-published prices.
Entry paid plan
Sim
Pro plan at $25/user/monthPro: $25 per user/month
Vellum
~$500/mo reported by a third party, unconfirmed by VellumUnconfirmed: a third-party analysis reports a Pro tier around $500/month for the enterprise platform, but Vellum does not publish this figure itself. The cited third-party breakdown describes Pro at roughly $500/month with 5,000 prompt executions/day, 250 workflow executions/day, RBAC, and monitoring integrations, still capped at 5 users; Enterprise tiers above that are custom/'Contact us' pricing. None of this is confirmed on Vellum's own site, which currently shows only the unrelated consumer product's $50/mo Pro plan.
Free tier
Sim
Vellum
Reported by a third party, unconfirmed by VellumReportedly yes: a third-party analysis describes a Free tier (50 prompt executions/day, 25 workflow executions/day, up to 5 users, no RBAC) for the enterprise platform, but this is not confirmed on Vellum's current site. Vellum's own pricing page no longer shows this tier; it now presents only the unrelated consumer 'Personal Intelligence' product's free 'Base' plan.
Bring your own key
Sim
Vellum
Unverifiable after the pricing page pivoted to the consumer productUnconfirmed for the enterprise platform: Vellum's pricing page has been replaced by an unrelated consumer 'Personal Intelligence' product page, so BYOK availability can no longer be verified from vellum.ai/pricing.. The current pricing page describes Vellum-provided credits for the consumer product, not the enterprise platform's LLM-cost pass-through model; no bring-your-own-API-key policy is described for either product on this page.
Security & compliance
SOC 2
Data residency
Sim
Full control via self-hosting; Cloud region toggle is global, not per-customerFull data control via self-hosting (Docker/Kubernetes); data never leaves customer infrastructure when self-hosted. On Sim Cloud, async job execution has an internal US/EU region toggle, but it is deployment-wide, not a customer-selectable per-workspace residency option
Vellum
No selectable residency regions documentedUnknown: no specific region/residency options documented. Docs describe data being stored 'in Vellum's infrastructure, isolated in a dedicated, encrypted container' but do not specify selectable residency regions.
Role-based access control
Sim
Yes: admin/write/read workspace permissions, org-level admin/member roles, plus Enterprise-tier permission groups that allow/deny-list specific models, tools, and integrations per group on top of those base roles. See modelAndToolGovernance and credentialGovernance for the finer-grained permission-groups layer.
Vellum
Yes: workspace-level role-based access control with six predefined roles. Roles are Admin, Deployment Editor, Document Index Editor, Test Suite Editor, Playground Editor, and Member (read-only). Permissions apply workspace-wide rather than per individual resource, and only Admins can change other users' roles.
Audit logging
Sim
Vellum
Not publicly documentedUnknown. Not publicly documented for the enterprise platform; permissions-model docs describe per-action risk badges for the consumer assistant product only.
Additional compliance
Sim
SOC2SOC2. Self-hosting is the primary lever Sim offers for data-residency-sensitive compliance needs beyond SOC2, rather than additional certifications.
Vellum
HIPAA + BAA; no other certs confirmedHIPAA compliant, with a BAA available for enterprise customers; no other certifications (ISO 27001, GDPR-specific attestation, PCI, FedRAMP) confirmed. Vellum's docs and a third-party Drata case study state it is HIPAA compliant and that enterprise customers can sign a Business Associate Agreement (BAA). No mention of ISO 27001, PCI, or FedRAMP certification was found.
Model & tool governance
Sim
Yes: enterprise "permission groups" let an admin allow-list/deny-list specific LLM providers and models, and separately deny specific tools/integrations (or disable all MCP or custom tools) per group, layered on top of workspace admin/write/read roles. This does not control whether an LLM provider retains prompts. Sim offers no "zero data retention" mode or governed AI gateway. A separate, Enterprise-gated feature lets orgs set a log-retention window and redact PII, but that only controls how long Sim itself keeps execution logs.
Vellum
Not publicly documentedUnknown. Not publicly documented.
Credential governance
Sim
Yes: shared credentials (connected accounts, service accounts, workspace secrets) are their own nested permission level (Member/Admin) below organization and workspace roles, and enterprise permission groups can further allow-list specific integrations and restrict which file-share auth modes (public/password/email/SSO) a group may use. A user's personal environment variables/secrets are never shared or inherited by anyone, including org owners/admins.
Vellum
No: Vellum has workspace-level Role-Based Access Control (Admin/Member roles governing create/update/delete permissions), but nothing documents restricting which specific stored credentials or connections a role or permission group may use.. RBAC docs describe workspace-wide role permissions (Admin vs Member) rather than credential-level allow/deny lists. A separate Vellum personal-assistant product describes per-credential allowedTools/allowedDomains scoping, but it is a different product from the B2B workflow platform compared here.
Single sign-on (SSO)
Sim
Vellum
Claimed by third parties, undocumented directlyUnknown: third-party summaries claim Vellum supports SSO/SAML, but no first-party Vellum documentation describes SAML/OIDC setup or auto-provisioning on first login.. Vellum's security/data-privacy documentation does not mention SSO/SAML, and a search of docs.vellum.ai for SSO/SAML configuration returns no dedicated setup page.
Vetted first-party integrations
Sim
Yes: every one of Sim's 266 blocks is first-party authored and code-reviewed through the standard pull-request process in the main Sim repository; there is no public marketplace where an arbitrary third party can publish and have other users install executable tool code without going through Sim's own review. Custom code steps run inside Sim's own isolated-vm sandbox rather than as an installable third-party skill package, so the supply-chain trust boundary is Sim's codebase review, not an open registry.
Vellum
Yes: Vellum's tool ecosystem is closed and vendor/partner controlled, not an open marketplace. Its 100+ native integrations are built and maintained by Vellum, its Composio partnership adds access to Composio's curated tool library, and 'Custom Nodes' are authored by the customer's own team for internal reuse rather than published to a shared marketplace for other tenants.. No documentation describes a public marketplace or community-node/plugin registry where third-party developers publish executable code for other Vellum customers to install, unlike ecosystems such as n8n community nodes. Custom Nodes extend a single customer's own workflows and are not distributed to other organizations. No publicly documented security incidents involving Vellum's integration or tool ecosystem were found.
PII redaction
Sim
Vellum
UnknownUnknown: nothing confirms or denies a dedicated PII detection/redaction feature distinct from generic Guardrail nodes for output validation.. Vellum documents a general Guardrail Node for workflow quality checks, but no page specifically describes PII detection/redaction (emails, SSNs, etc.) in workflow content or logs.
Custom data retention
Sim
Vellum
Yes: Enterprise customers can configure data retention policies to automatically delete monitoring/interaction data after a specified period (30, 60, 90, or 365 days) instead of the default indefinite retention.. Configured from Organization Settings under Advanced Settings; default behavior without this Enterprise configuration is indefinite retention.
White-labeling
Sim
Vellum
Unknown, no white-label branding docs foundUnknown: no documentation describes replacing Vellum's own branding (logo, product name, theme) with a customer's branding across the workspace or deployed apps. Only generic 'white glove' Enterprise service language appears, and that refers to onboarding support, not white-label branding.. Vellum's own branding-guide page addresses Vellum's use of its own brand assets by others, not a customer-facing white-label capability.
AI capabilities
Multi-LLM support
Sim
21 providers plus dynamic-resolution aggregators (OpenRouter, LiteLLM, etc.)21 provider integrations (OpenAI, Anthropic, Google/Gemini, Azure OpenAI, Azure Anthropic, Groq, Cerebras, Mistral, xAI, Bedrock, Vertex, Ollama, OpenRouter, and more), with OpenRouter, LiteLLM, vLLM, and Ollama resolving models dynamically at runtime rather than from a fixed list, so effective model reach extends well beyond the 21 named providers. apps/sim/providers/models.ts defines 21 provider entries.
Vellum
20+ providers, hundreds of modelsModel-agnostic across 20-23+ providers (OpenAI, Anthropic, Google/Gemini, Cohere, Azure OpenAI, Bedrock, Fireworks, Perplexity, Cerebras, Groq, etc.) and hundreds of individual models, including recent additions like GPT-5 and Claude Opus 4.1.
Agent reasoning blocks
Sim
Vellum
Agent Node handles reasoning + tool execution; MCP not currently documentedVellum's 'Agent Node' (formerly 'Tool Calling Node') is its dedicated agent/reasoning-and-tool-execution block within Workflows, currently documenting raw code, subworkflows, and Composio SaaS actions as tool types in one node; MCP tool support is not documented on the current Agent Node page.
Natural-language building
Knowledge base / RAG
Sim
Yes: hybrid vector (pgvector) plus full-text (tsvector) search knowledge base, 11 supported file formats (csv, doc, docx, html, json, md, pdf, pptx, txt, xlsx, yaml), configurable chunking, plus 51 connectors that continuously sync external sources (Google Drive, Confluence, Slack, Gmail, GitHub, HubSpot, Linear, Jira, and more) into the knowledge base rather than a one-shot upload
Vellum
Document Indexes + RAG pipeline evaluationVellum supports RAG via 'Document Indexes' and a dedicated 'Evaluating RAG Pipelines' documentation area.
MCP support
Sim
Vellum
Remote MCP servers with auto tool discoveryAgent Nodes support adding a remote MCP server as a tool via a '+ Tool' button, with automatic tool discovery (since August 2025).
Evaluation & guardrails
Sim
LLM-judge Evaluator plus Guardrails validation blockEvaluator block (LLM-judge scoring against user-defined named metrics) and Guardrails block (JSON validity, regex, RAG/hallucination scoring, PII detection/masking). These are per-call scoring/validation primitives, not a batch golden-dataset eval-suite runner or A/B prompt-testing harness.
Vellum
Test suites, scored metrics, production evalsVellum's 'Evaluations' product uses Test Suites and Metrics (0-1 scores) for quantitative evaluation, supports bulk CSV test-case upload, online/production evaluations, custom reusable metrics, and RAG-pipeline evaluation.
Human-in-the-loop
Sim
Vellum
External Input node for approvalsAn 'External Input' node pauses Workflow execution until a human or external system supplies input, enabling human-in-the-loop approval patterns.
Generative media
Sim
Vellum
Image-gen credits exist; blocks undocumentedUnknown. Pricing includes credits for image generation as a billable usage type, but specific image/video/audio blocks or supported providers are not publicly detailed.
Dynamic tool use
Sim
No: an Agent block calls tools the workflow author explicitly added to it at build time, rather than browsing and picking from a broader pool (e.g. an entire MCP server catalog) at inference time. Runtime MCP "discovery" exists to resolve/refresh the schema of an already-configured tool. The model does not browse or choose from the server's full tool list.
Vellum
Not publicly documentedUnknown. Not publicly documented.
Automatic model fallback
Sim
No: a failed or rate-limited LLM call is retried using Sim's own hosted API keys for the same model, rather than automatically switching to a different model or provider. A "fallback" comment in the provider layer refers to rotating among Sim's own hosted API keys for the same model, not switching models.
Vellum
Not publicly documentedUnknown. Not publicly documented.
Agent skills
Sim
Yes: named, reusable "Agent Skills" (built on the open Agent Skills / SKILL.md format) that agents load on demand via progressive disclosure, editable in-app or imported from a SKILL.md file or GitHub URL. Only the skill name and description sit in the agent's system prompt (~50-100 tokens each); the full instructions load into context only when the agent calls load_skill.
Vellum
No: Vellum's B2B workflow/agent platform (docs.vellum.ai) has no named, reusable prompt or skill object invoked by reference across multiple agents. Its reuse primitives are Subworkflows (reusable workflow logic blocks) and shared Prompt Sandboxes, not a discrete 'skill' package referenced by name.. Reuse happens via Subflows/subworkflows and deployed prompts, a different mechanism than a discrete named skill library referenced across agents. Vellum's separate consumer 'Personal Intelligence' assistant product (vellum.ai/docs, a different product from the B2B workflow platform compared here) does have a 'Skills' feature, built on the same open convention as other agent platforms: a directory containing a SKILL.md file (YAML frontmatter plus a markdown instruction body, with optional bundled scripts/reference files), extended with Vellum-namespaced metadata fields like metadata.vellum.display-name.
Native chat deployment
Sim
Vellum
Yes: Vellum Agents can be built with a first-class Chat Message Trigger that maintains chat_history/conversation state across turns, and Workflows/Agents can be deployed with this chat pattern rather than only form/API/webhook targets.. Documented alongside RAG chatbot tutorials and the Agent Node's conversation-state handling; deployment surface details (e.g., a hosted public chat widget URL) were not independently confirmed beyond the trigger/state mechanism.
Parallel execution
Sim
Yes: a native Parallel block fans a run out into concurrent branches (fixed count or one per list item) and joins their results back into the workflow automatically. Contained blocks run concurrently instead of sequentially, either a fixed number of times or once per item in a list/collection, and each branch's output aggregates for downstream blocks.
Vellum
Yes: Vellum Workflows has a Map Node that iterates over an array and executes a subworkflow concurrently for each item, and a Merge Strategy (available on all node types, replacing the older standalone Merge Node) that consolidates divergent execution paths back into one result.. As of the January 2026 release, the standalone Merge Node was replaced by Merge Strategy, a setting on every node type, so branches can fan out via the Map Node and fan back in without a dedicated join node.
Agent2Agent (A2A) protocol
Sim
Vellum
No: Vellum documentation and changelogs show no support for the Agent2Agent (A2A) protocol. Vellum has written about the related Google AP2 payments protocol, but has not documented an A2A implementation or Agent Card support.. No mentions of "Agent2Agent" or "A2A" appear in Vellum product docs, help center, or changelog. This reflects the absence of public documentation, not a statement from Vellum that it will never support it.
Loop / iteration block
Sim
Vellum
No: Vellum's only documented list-iteration mechanism is the Map Node, which executes a subworkflow once per array item concurrently (up to 96 parallel executions) rather than as a dedicated sequential for-each/while container; no separate While/loop node is documented.. The Map Node is already the mechanism counted under parallelExecution (concurrent fan-out plus Merge Strategy join). Vellum does not document a way to force single-lane sequential iteration or a distinct while/repeat-until construct.
Integrations
Integrations
Sim
1,000+ integrations (266 blocks, ~3,900 tool actions)1,000+ integrations counting individual API actions, built from 266 first-party blocks and roughly 3,900 underlying tool actions. Sim's landing page cites the "1,000+ integrations" figure; the block/tool-action counts are the same integration surface measured at a different level of granularity.
Vellum
10,000+ tools via Composio, 100+ nativeVia a Composio partnership (August 2025), Vellum connects to Composio's library of 10,000+ tools directly inside Agent Nodes (Google Sheets, Slack, Salesforce, Notion, Jira, Linear, Trello, etc.). Vellum separately advertises 100+ of its own native integrations.
Trigger types
Sim
Webhook, cron, chat, REST API, 61 app triggers, plus a diff-aware Tables triggerWebhook, schedule/cron, chat, REST API, and event-based triggers for 61 apps (Slack, Gmail, GitHub, Stripe, etc.), plus a native row-level trigger on Sim Tables that fires on insert/update with an optional column watch-list and emits a before/after diff
Vellum
Chat trigger confirmed; full list undocumentedUnknown. A 'Chat Message Trigger' exists for chat-first agent building; the full set of trigger types (webhook, API, etc.) is not publicly enumerated.
Custom code steps
Vellum
Python/TypeScript code node with in-browser IDEVellum has a documented 'Code Execution Node' supporting custom Python or TypeScript code with a required main() function signature, an in-browser IDE, and support for public PyPI/npm packages. Newer 'Custom Nodes' are expected to eventually replace it.
API publishing
Sim
Yes: a public REST API (mostly under /api/, with /api/v1 reserved for logs and audit-log endpoints) supporting API-triggered workflow execution and deployment rollback. The API reference does not document SSE streaming, a resumable event buffer, a dedicated API-trigger block, or a chat-deployment surface as part of the REST API itself.
Vellum
One-click deploy to a callable APIDeploying a Workflow from the Vellum UI produces a code snippet to call it in production as an API. Vellum handles execution server-side and callers just supply input variables; each execution is also viewable/shareable via an execution URL.
SDKs & extensibility
Sim
Official Python and TypeScript SDKs, plus MCP, code block, and A2A protocolOfficial client SDKs exist for Python (simstudio-sdk, pip install simstudio-sdk) and TypeScript/JavaScript (simstudio-ts-sdk, npm install simstudio-ts-sdk), both wrapping the REST API (x-api-key header) with methods like execute_workflow / executeWorkflow, retry-with-backoff, and usage-limit lookups. Extensibility is further extended by MCP (client + server), a sandboxed code-execution block (JS/Python), custom tools, and an Agent-to-Agent (A2A) protocol block for external agent interop
Vellum
Open-source Python Workflows SDKThe Workflows SDK (GitHub: vellum-ai/vellum-python-sdks) is an open-source Python framework for defining and executing agentic workflows as graphs declaratively; docs also describe a 'Custom Nodes' extensibility tutorial.. Language support beyond Python is not publicly confirmed.
Publish as MCP server
Sim
Vellum
No: Vellum's MCP support runs one direction only. Its Agent Node lets a workflow connect to and call external/remote MCP servers as tools (with auto-discovered schemas), but nothing documents the reverse: publishing a deployed Vellum workflow as a callable MCP server for external AI tools to consume.. The August 2025 changelog and blog content describe adding MCP servers as tools inside Agent nodes; Vellum's 'How does MCP work' blog post does not address exposing Vellum workflows as MCP endpoints. Some third-party sources conflate this with Vellum's separate personal-assistant product exposing its own MCP server, a different, non-workflow-platform product.
Observability & durability
Tracing & observability
Sim
Yes: execution logs include a per-block/per-span trace view (duration, cost, token counts, and latency stats like TTFT/TPS) with expandable nested iteration groups, plus a "View Snapshot" frozen copy of the workflow structure and block states at run time for debugging. This trace view is built directly into Sim rather than a raw export browsable in an external tool like Jaeger, and does not expose aggregate latency-percentile charts (p50/p95/p99). The run snapshot serves as a log-detail/debugging artifact rather than a resumable mid-run checkpoint.
Vellum
Full execution tracing with shareable URLsVellum automatically captures every Prompt/Workflow execution in production into filterable, sortable Executions tables, including per-step inputs, outputs, latency, and aggregated cost, plus shareable execution URLs for linking from external tools and alerts.
Durability & retries
Sim
Tool-call retries (up to 10x); single-attempt job orchestrationIndividual tool/API calls have configurable exponential-backoff retry (up to 10 attempts). The background job-orchestration layer itself retries only once by design. Durability instead comes from consecutive-failure tracking on schedules and the human-in-the-loop snapshot pause/resume mechanism. Sim does not offer guaranteed-once-only block execution, a failed-run holding queue for manual recovery, or a "replay a past execution with its original inputs" feature. The per-execution debugging snapshot serves as a log-detail artifact rather than a resumable mid-run checkpoint.
Vellum
Automatic node-level retriesVellum supports a 'Retry Node Adornment' — a standalone adornment applied to a node (separate from that node's Settings) that automatically re-invokes the node until it succeeds or reaches a configured max-attempts count.
Failure alerting
Sim
Vellum
Not publicly documentedUnknown. Not publicly documented.
Data drains
Sim
Yes: Enterprise orgs can continuously export workflow logs, job logs, or audit logs on a schedule to a customer-owned S3 bucket, GCS bucket, Azure Blob container, BigQuery table, Snowflake table, Datadog logs intake, or an HTTPS webhook. Each drain exports exactly one data source; multiple drains are created to export multiple sources. Viewing drain config/run history is restricted to org owners/admins.
Vellum
Yes: Vellum streams execution/monitoring events (workflow execution initiated/fulfilled/rejected, usage calculation, metric execution events) continuously to external systems via configurable webhooks, including documented support for forwarding to Datadog.. Configured from Organization Settings; supports API key, Bearer token, and HMAC verification for webhook payloads. No native S3 or BigQuery connector exists, but the generic webhook mechanism supports building such an export.
Async execution
Sim
Yes: a workflow can be triggered in fire-and-forget async mode, returning HTTP 202 with a job ID immediately, then polled via a dedicated jobs endpoint through queued/processing/completed/failed states. Async jobs are tracked via polling the job endpoint rather than a completion webhook/callback option.
Vellum
Yes: Vellum offers an Execute Workflow Async API endpoint that starts a workflow run and returns an execution_id immediately, without blocking. A separate status endpoint lets clients poll for the current state (PENDING, FULFILLED, REJECTED, etc.) and outputs once the run finishes.. Introduced November 2025. Vellum also documents an Execute Workflow as Stream endpoint for streaming responses, in addition to the synchronous Execute Workflow call that blocks until completion.
Execution limits
Sim
5-50 min sync timeout, 90 min async, 10-1,000 concurrentPlan-gated: synchronous API calls time out at 5 minutes on the free plan and 50 minutes on paid plans, async calls at 90 minutes on every plan, with 10 to 1,000 concurrent executions per billing account depending on plan. Concurrency limits are published in the platform cost docs, and Enterprise limits are customizable. Request bodies are separately capped at 10 MB.
Vellum
Concurrency limit exists, no public numbersUnknown: Vellum confirms a per-account concurrency limit exists and that async executions automatically queue once it is exceeded, but does not publish the actual numbers. No max concurrent executions, max execution duration, or requests-per-minute rate limit is listed on its public docs or pricing pages.. The November 2025 changelog states executions 'automatically queue when you exceed your concurrency limit' but gives no figure. The pricing page describes credit-based billing and machine sizes (vCPU/RAM tiers) but no execution timeout or concurrency figures. Third-party blog posts cite older per-day execution caps, but Vellum's current docs do not confirm them, so they are not included as a verified figure.
Partial-failure handling
Sim
Vellum
Yes: Vellum lets you wrap any workflow node with a Try or Retry adornment for first-class error handling, so a single node's failure does not have to halt the entire run. The Try adornment attempts the node once and continues with an Error output if it fails; the Retry adornment repeatedly re-invokes the node until it succeeds or hits a maximum attempt count.. Adornments are applied from the node's side panel and appear in monitoring as a single-node subworkflow, so downstream branches can consume the Error output and keep the rest of the run going instead of the whole execution stopping.
Unattended execution
Sim
Yes: scheduled, webhook, and chat-triggered runs execute as background jobs on trigger.dev workers, entirely on Sim's servers. No client device needs to be open, awake, or connected for a run to fire or complete; closing the browser tab or shutting down a laptop has no effect on a scheduled or triggered workflow.
Vellum
Yes, by inference: Scheduled Triggers (cron-based) and Integration Triggers (webhook-based, introduced November 2025) fire a deployed Workflow automatically on Vellum's own Cloud, self-hosted, or VPC infrastructure, the same server-side execution path already documented for API and async calls. No Vellum documentation for the B2B workflow/agent platform ties a scheduled or triggered run's reliability to a client device, browser tab, or desktop app staying open.. Vellum's changelog states a deployed Workflow Deployment containing a Trigger 'executes automatically based on that configuration' but does not spell out the runtime location in those words. This is inferred from the deployment/API execution model (deploymentOptions, apiPublishing, asyncExecution facts) rather than an explicit doc statement. Note Vellum's separate consumer 'Personal Intelligence' assistant product explicitly ties its own schedule reliability to a locally running daemon on self-hosted installs; that client-dependent model is a different product from the B2B workflow platform compared here.
Support
Support channels
Sim
Community support plus Enterprise 'Dedicated Support'Community (open source, GitHub) plus an unquantified "Dedicated Support" flag on the Enterprise plan. Enterprise and pricing pages do not include CSM, onboarding/enablement, or professional-services details beyond a plan-comparison-table "Dedicated Support" flag.
Vellum
Email, in-app chat, Slack (customers), Discord communityEmail, an in-app chat, a shared Slack channel for active customers, and a Discord community. Vellum's help center lists email (support@vellum.ai), in-app chat via the dashboard's 'Get Help' button, and a shared Slack channel for active customers; the enterprise page separately links a public Discord community. vellum.ai/pricing has since been repurposed for a different 'personal AI assistant' product, so its 'priority support' mention describes that product's paid add-on, not the workflow platform's support tiers — no priority-support or SLA claim is made here.
SLA
Sim
Vellum
No SLA content found on current siteUnknown: no SLA commitments or figures found on current public pages. The enterprise and pricing pages now serve the consumer 'Personal Intelligence' product and contain no SLA language (uptime, response time, or otherwise). An earlier version of this page may have referenced named SLA features, but the live site does not confirm that, so it is not included as a verified claim.
Community
Sim
100,000+ buildersOver 100,000 builders use Sim
Vellum
Discord communityDiscord community exists
Academy / training
Sim
Vellum
No: Vellum has no documented Academy-style learning resource with courses or certification. It offers standard documentation (docs.vellum.ai), a blog, and webinars, but no dedicated course or certification program.. Searches for 'Vellum academy', 'certification', 'courses' turned up only docs, blog posts, and webinars; no evidence of a structured curriculum.
Sim standout features
AI Copilot / Chat agent-building surface
Chat builds and manages work across the workspace; in-editor Copilot edits a single workflow.
A workspace-wide natural-language surface (Chat) that can build workflows, manage data, and take actions across integrations, plus an in-editor Copilot scoped to building and editing a single workflow directly.Hybrid semantic + keyword knowledge base
Combines vector and full-text search with configurable chunking across 11 file formats.
Built-in RAG with pgvector embeddings and a generated tsvector column for combined vector + full-text search, plus a token-based chunker with configurable chunk size/overlap and 11 supported file formats (csv, doc, docx, html, json, md, pdf, pptx, txt, xlsx, yaml).Native MCP client and server
Call external MCP servers as tools, or expose Sim workflows as an MCP server.
A dedicated MCP block lets any workflow call external MCP servers as a tool, and a serve/workflow-servers API surface lets Sim expose its own workflows as MCP servers.Fork a workspace into dev, qa, and prod environments
Fork, diff, and promote environments with mandatory credential remapping.
Fork a whole workspace into a dev/qa/prod-style child environment, preview a diff, and promote changes bidirectionally. Credential and env-var remapping is required on every promote, so secrets never cross environments silently.Human-in-the-loop approvals with durable resume
Pause a run for human approval and resume later via a durable snapshot link.
A dedicated block pauses a run and waits for a human-submitted approval form, backed by persisted execution snapshots so the run can resume later via a link, even after a server restart.Self-hostable under Apache 2.0
Fully open source with Docker Compose and Helm deployment options.
Fully open source (Apache 2.0), with Docker Compose files and a Helm chart for Kubernetes deployment, alongside a managed cloud-hosted option.Live multiplayer canvas editing
Real-time cursors, selections, and synced edits on the same canvas.
Real-time cursors, selection broadcasting, and synced concurrent edits over a dedicated realtime backend, so a team can build the same workflow together at the same time.Documented Vellum limitations
Brand/product ambiguity as of mid-2026
Public site now foregrounds a consumer product over the enterprise platform.
Vellum's homepage, pricing page, security docs, and several product pages now serve content for a new consumer 'Personal Intelligence' assistant (an open-source, MIT-licensed Mac app) instead of the original enterprise workflow, evaluations, and prompt-engineering platform. That makes enterprise-specific details, like environment promotion, version-control depth, tracing, alerting, and integration counts, harder to verify from the public site alone.Enterprise BYOK policy no longer verifiable
Pricing page now shows a different consumer product; enterprise BYOK policy unconfirmed.
Vellum's pricing page has been replaced by a consumer 'Personal Intelligence' pricing model (subscription plus pay-as-you-go credits); the original enterprise LLM-cost-pass-through pricing this claim describes, including any bring-your-own-API-key policy, is no longer published at this URL.Enterprise SLA no longer verifiable
SLA status unverifiable after the /enterprise page pivoted to the consumer product.
The /enterprise page has been repurposed for Vellum's consumer product and no longer represents its (formerly documented) enterprise LLMOps offering, so SLA absence can no longer be verified against genuine enterprise-tier content at this URL.No native spreadsheet-style data tables
No documented table/spreadsheet feature; CSV upload and extraction only.
No Vellum documentation describes a native, spreadsheet-like data-table object with persistent rows/columns, keyboard navigation, or per-row writes from workflows; the closest documented tabular handling is uploading CSV/XLS files and extracting or generating structured output within workflow nodes, unlike Sim, which has a built-in Tables feature.Bottom line
Choose Sim if you want an open-source, self-hostable AI workspace that treats AI agents as first-class citizens: native multi-LLM support, real-time multiplayer editing, environment promotion (dev/qa/prod), human-in-the-loop approvals, and enterprise governance (SSO, credential-level permissions, audit logs) built in rather than bolted on.
Choose Vellum if you specifically need HIPAA compliance with a signable BAA: Vellum is HIPAA compliant and enterprise customers can sign a Business Associate Agreement for handling protected health information, corroborated by a third-party Drata case study. Sim does not currently offer HIPAA compliance (both platforms hold SOC 2 Type 2).
Frequently asked questions
Sim is an open-source AI workspace where teams build, deploy, and manage AI agents visually, conversationally, or with code. Vellum is an enterprise AI development platform for building, evaluating, and deploying LLM prompts, workflows, and agents. Teams considering a switch typically weigh licensing (Sim is Apache 2.0 and self-hostable), pricing model, and how AI-native the platform's agent-building experience is.