Skip to content

Story continuation guide

Continue a comic from an existing generation.

The continuation operation sends action continueWrite to an existing generation ID so LlamaGen can extend that comic instead of starting a disconnected job from a new prompt.

Continuation reuses generation context, but it is not durable series memory or automatic publication. Keep the approved canon, issue plan, and continuity record in your own system and review every returned page.

Review the continuation contract

Account, API key, usage, and billing open on LlamaGen.AI.

Direct answer

The decision this page is for.

Last verified:

Continue an existing comic by PATCHing its generation ID with action continueWrite, a next-story prompt, and an explicit page or panel bound. The existing ID anchors the operation, but it is not durable series memory: store the canon yourself, track asynchronous state, and review the new pages before release.

Verifiable request

Continue an existing generation

cURL
curl -X PATCH \
  https://api.llamagen.ai/v1/comics/generations/gen_abc123 \
  -H "Authorization: Bearer $LLAMAGEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "action": "continueWrite",
    "prompt": "The courier follows the amber signal into the closed station.",
    "pagination": { "totalPages": 1, "panelsPerPage": 4 }
  }'

Structured response

State your application can route

response.jsonILLUSTRATIVE
{
  "id": "gen_abc123",
  "status": "LOADING"
}

Continuation walkthrough

Continue the same story without hiding the editorial handoff.

The studio view places prior pages, recurring characters, the next sequence, and creator review in one observable workflow.

An independent creator compares an existing comic issue with newly continued pages pinned in a working studio
Editorial illustration generated with AI; it is not a creator photograph and not a generated API result.Art direction · Independent comic-studio documentary

Continuation control

Anchor the request to an approved generation and an external canon.

A useful continuation flow records which generation was approved, what the next story beat should accomplish, and who accepts the new pages.

  1. 01Creator

    Freeze the accepted state

    Store the approved generation ID, exported pages, bible version, and unresolved story beat.

  2. 02Server

    Request continuation

    PATCH the exact generation ID with continueWrite, the next story beat, and an explicit page or panel bound from a trusted runtime.

  3. 03Application

    Track the lifecycle

    Show LOADING state and handle documented error responses instead of waiting in one browser request.

  4. 04Creator or editor

    Compare the pages

    Check story logic, recurring characters, props, locations, captions, and reading order against the saved canon.

  5. 05Publishing workflow

    Accept or revise

    Archive accepted pages and selectively repair failures before assembly or release.

Capability and boundary facts

Facts an integration can verify.

Last verified:

Operation
PATCH /v1/comics/generations/{generationID}
Action
continueWrite with a next-story prompt
Request bound
pagination or fixPanelNum
Context anchor
Existing generation ID
Documented failures
400, 402, 403, 429, and 500 responses
Verification
LlamaGen documentation and OpenAPI checked on 2026-10-28

Good fit

Use it when

  • Adding a next segment to an approved generation
  • Serialized creator workflows with an external series bible
  • Reviewing continuity before assembling an issue

Wrong fit

Do not use it for

  • Assuming the service remembers unrelated jobs
  • Generating an entire series without review gates
  • Publishing continuation output automatically

Production limits

Plan for the boundaries.

  • 01Continuation is probabilistic; it does not guarantee plot, wardrobe, lettering, or character continuity.
  • 02Use pagination or fixPanelNum to bound the requested continuation, then verify the returned structure before committing a fixed deliverable to a user.
  • 03A PATCH response with LOADING still represents ongoing work. Track the generation until PROCESSED or FAILED.
  • 04The example response is contract-shaped placeholder data, not the result of a request executed from this page.
  • 05Generation is asynchronous; production interfaces must handle processing and failure states.
  • 06Outputs should be reviewed against product, brand, safety, and print requirements before delivery.
  • 07You must have the rights to prompts, photos, scripts, and reference assets submitted to the API.

Authoritative sources

LlamaGen Comic API documentationLlamaGen OpenAPI contractCurrent pricingTermsService status

Questions

Direct answers before you integrate.

01Which endpoint continues a comic?
PATCH /v1/comics/generations/{generationID} with action: continueWrite, a next-story prompt, and either pagination or fixPanelNum when you need an explicit size. Authenticate from a trusted server.
02Will continuation preserve every character and plot detail?
No. Continuation reuses generation context but remains probabilistic. Compare characters, wardrobe, props, locations, lettering, and plot logic against your saved series bible.
03How do I bound the continuation size?
Use pagination to request pages and panels, or fixPanelNum for a fixed panel count. Those fields bound the request; the generated story, composition, and continuity still need review.

Related evaluation paths

Use caseComic API for Indie Creators and Serialized ComicsFormatStory-to-Comic API for Scripts and Story BriefsFormatCharacter Consistency API for Multi-Panel Comics

Next step

Move from evidence to a real API request.

LlamaGen provides the playground, API key, usage, billing, and authoritative service documentation.

Review the continuation contract