Save 4 hours! We auto-setup your workflow! Free!

Drop your workflow.json — we handle every dependency, custom node, and model. Just open the link and run.

Auto-Setup Workflow Json (Free) Now!
ComfyUI > Nodes > MiniMaxH3-Studio > H3 Source-Profiled Renderer

ComfyUI Node: H3 Source-Profiled Renderer

Class Name

comfyui_h3_context.H3Context.SourceProfiledRenderer

Category
h3_context/compiler
Author
rookiestar28 (Account age: 1077days)
Extension
MiniMaxH3-Studio
Latest Updated
2026-11-05
Github Stars
0.16K

How to Install MiniMaxH3-Studio

Install this extension via the ComfyUI Manager by searching for MiniMaxH3-Studio
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter MiniMaxH3-Studio in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

H3 Source-Profiled Renderer Description

Renders H3 context plans into prompts and validates them against a pinned official guide.

H3 Source-Profiled Renderer:

The H3 Source-Profiled Renderer turns an accepted H3 context plan into prompt text and checks that it conforms to a pinned official H3 guide. Here, “rendering” means preparing and validating a prompt, not generating an image, video, or audio file. The node first validates the complete plan, then selects the official guide reference appropriate to its prompt profile: the full-reference guide for the full-reference profile, or the base guide otherwise. It binds that profile to the guide’s recorded revision and digest, a fingerprint used to identify the expected source content, before preparing the prompt and checking its validity. This gives you a controlled, source-grounded prompt preparation step rather than unrestricted text formatting. If the plan or resulting prompt fails the required checks, the node stops instead of passing an unvalidated result onward. Its main benefit is helping you keep prompt preparation consistent with a specific official reference while preserving structured results for later workflow stages; it does not establish that a media-generation engine is available or that generated media will satisfy the prompt.

H3 Source-Profiled Renderer Input Parameters:

plan

Required input of type H3_CONTEXT_PLAN containing the structured H3 context plan to prepare and validate. You must connect an actual ContextPlan object from a compatible upstream workflow; plain text, a dictionary, or even a subclass of that object will not pass the node’s exact-type check. The node validates the complete immutable plan structure before reading its nested contents, so a connection with the correct socket type alone does not guarantee acceptance. The plan’s request profile determines which pinned official guide is used: PromptProfile.FULL_REFERENCE selects the full-reference guide digest, while other accepted profiles follow the base-guide branch. The plan therefore controls both the prompt content to be processed and the reference against which it is checked. Minimum: not applicable. Maximum: not applicable. Default: none specified. Options: no selectable options are exposed on this node; profile selection belongs to the upstream plan.

H3 Source-Profiled Renderer Output Parameters:

prompt

Output of type H3_PROMPT_STRING containing the text of the successfully validated prompt document. This is the text-oriented result to use wherever a compatible downstream node expects an H3 prompt string. It is taken directly from the same document returned through prompt_document, so these two outputs are not separately generated prompt variants. The node returns this text only after the source-profiled result passes its validity check and its registration step completes. Interpret it as a validated prompt prepared under the selected source contract, not as generated media or evidence that a media renderer has executed.

profiled_result

Output of type H3_SOURCE_PROFILED_PROMPT containing the structured SourceProfiledPromptResult produced by source-profiled prompt rendering. This preserves the result as a typed object rather than reducing it to text, making it suitable for downstream stages that explicitly accept this socket type. The node checks that the result’s is_valid value is true before returning it, and its document supplies both the text and document outputs. This output is important when your workflow needs the structured source-profiled result for further processing instead of only the written prompt. The supplied code does not define its complete field layout, so additional metadata fields should not be assumed.

prompt_document

Output of type H3_PROMPT_DOCUMENT containing the PromptDocument associated with the accepted source-profiled result. Use it when a downstream node requires the prompt as a structured document rather than a text string or the complete profiled result. Its text property is exactly what the node returns through prompt, which keeps the textual and document-based branches of your workflow aligned. This output represents the successfully validated document; it is not a separate validation report, generated image, or rendered video. The supplied interface does not describe any further document fields.

H3 Source-Profiled Renderer Usage Tips:

  • Connect plan to a compatible H3 plan-producing stage. Do not try to replace the structured plan with pasted prompt text, even if that text describes the same scene.
  • Choose your intended prompt profile upstream. This node has no local profile selector: a full-reference plan selects the pinned full-reference guide, while other accepted profiles use the base-guide branch.
  • Route prompt to text-oriented stages, profiled_result to stages accepting H3_SOURCE_PROFILED_PROMPT, and prompt_document to stages accepting H3_PROMPT_DOCUMENT. Matching the downstream socket type avoids confusing a structured result with its text.
  • Treat validation failures as a reason to correct or regenerate the upstream plan. The node deliberately stops on failure rather than returning a partial or unchecked prompt.
  • Keep the related H3 custom-node components consistent when updating your installation. The workflow relies on exact object types and pinned source contracts; mixing incompatible components can prevent an otherwise plausible plan from being accepted.
  • Do not look for sampling, resolution, seed, or media-rendering controls here. The only exposed input is plan, so task-specific prompt choices must be made upstream, and actual media generation requires a separate compatible stage.

H3 Source-Profiled Renderer Common Errors and Solutions:

invalid_plan: plan must be an exact ContextPlan

  • Explanation: The supplied value is not exactly the ContextPlan type required by this node. A text string, dictionary, unrelated object, or subclass will be rejected before source-profiled rendering begins.
  • Solution: Connect a compatible H3 plan output directly to plan. Remove conversions that replace the plan with text or generic data. If the problem occurs between H3 nodes, check that they come from a consistent installation and regenerate the upstream plan.

source_profiled_render_failed: source-profiled rendering failed closed

  • Explanation: This is the node’s outer error for a caught ContractValidationError, TypeError, or ValueError during plan validation, source binding, prompt rendering, result validation, or registration. “Failed closed” means it stops rather than passing along a result that may violate the required contract.
  • Solution: Check the execution log for the underlying exception attached to this message. Correct the upstream plan or incompatible configuration identified there, then rerun the workflow. The outer message alone does not identify which check failed.

source-profiled prompt did not validate

  • Explanation: The rendering step returned a result whose is_valid value was false. The node raises this validation error internally and then wraps it in source_profiled_render_failed, so you may see it as the underlying cause rather than the main displayed message.
  • Solution: Review the upstream plan and its selected profile for compatibility with the pinned official guide. Use any additional validation details available in the execution log to revise the plan, then regenerate it rather than attempting to bypass the check.

invalid_contract: renderer profile must be typed

  • Explanation: The provided renderer-profile validation code rejects values that are not exactly RendererCapabilityProfile objects. This is a related contract-validation error, not an exposed setting on this node; if it propagates through rendering as a caught validation exception, the node reports its general source-profiled rendering failure.
  • Solution: Use the renderer capability profile supplied by the compatible H3 implementation instead of a manually constructed dictionary or unrelated object. If you have not customized profiles, check for inconsistent custom-node versions.

renderer_profile_unqualified: M25-17 cannot advertise renderer execution capability

  • Explanation: The related renderer-profile validator requires qualification to remain unqualified and execution capability to remain unavailable. It rejects a declaration that claims an available or qualified renderer. This restriction separates prompt preparation from verified media-rendering capability.
  • Solution: Restore the required unqualified renderer profile. Do not mark rendering capability as available merely because prompt preparation succeeds; use a separate media-generation workflow for execution.

renderer_profile_unqualified: unqualified renderer profile cannot carry invented implementation facts

  • Explanation: The related validator rejects an unqualified renderer profile if its renderer ID, version, build fingerprint, artifact fingerprint, or SPDX license identifier is not None. These fields cannot be populated to imply an implementation that this contract does not establish.
  • Solution: Use the required unqualified profile without adding renderer implementation details. If custom code populates those fields, remove that customization rather than inserting placeholder values.

unsupported_profile: renderer capability declaration drifted

  • Explanation: The related renderer-profile validator requires the profile to match the expected unqualified declaration exactly. A profile can therefore fail even after passing the basic type and execution-capability checks if another part of its declaration differs.
  • Solution: Restore the profile provided by required_unqualified_renderer_profile() in the compatible H3 implementation. If you are working only through ComfyUI, use a consistent custom-node installation and rebuild the affected upstream objects instead of editing their internal profile data.

H3 Source-Profiled Renderer Related Nodes

Go back to the extension to check out more related nodes.
MiniMaxH3-Studio
RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.

H3 Source-Profiled Renderer