Skip to content

docs(agent): document Spark 2, retire Spark 1 Pro and Mini - #1271

Open
rakshith48 wants to merge 1 commit into
mainfrom
docs/agent-spark-2
Open

docs(agent): document Spark 2, retire Spark 1 Pro and Mini#1271
rakshith48 wants to merge 1 commit into
mainfrom
docs/agent-spark-2

Conversation

@rakshith48

Copy link
Copy Markdown
Contributor

What

Agent runs on a single model now, Spark 2. spark-1-mini and spark-1-pro are removed from the docs as selectable models.

  • features/agent.mdx — Model Selection collapses to one spark-2 row; the Mini/Pro tip, the per-model "use when" sections, and the two-model parameter description are gone.
  • features/models.mdx — rewritten around Spark 2: one model, what improved versus Spark 1 (cheaper, faster, quality at least on par), pricing, and a note that the retired presets still run.
  • Snippets — agent/with-model/{python,js,curl} now show a single spark-2 call, agent/base/* sends spark-2, and the CLI options snippet uses --model spark-2.
  • advanced-scraping-guide.mdx, sdks/cli.mdx, developer-guides/usage-guides/choosing-the-data-extractor.mdx — model tables and inline comments updated.
  • api-reference/v2-openapi.json — request model enum is ["spark-2"] with spark-2 as the default. The response model field drops its enum, since archived runs still report the retired presets.

Each page states that spark-1-mini and spark-1-pro are retired but still accepted, so existing integrations know they need no code change.

Notes

  • Localized files (ja/, zh/, pt-BR/, fr/, es/, and their snippets) are untouched per the repo guideline — translations land through locadex.
  • The Parallel Agents pricing line mentioning Spark-1 Fast is left as is; that is a separate feature, not agent model selection.
  • Comparative claims are qualitative on purpose. Happy to swap in the concrete cost/latency multiples from the Spark 2 pricing analysis if we want them public.

Companion to firecrawl/firecrawl-web#3252, which makes the playground send spark-2.

🤖 Generated with Claude Code

Agent runs on a single model now. Model Selection collapses to spark-2,
the models page drops the two-model comparison, and every snippet, CLI
flag, options table, and OpenAPI enum sends spark-2.

Retired presets are noted as still accepted so existing integrations
know they need no code change.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
@mintlify

mintlify Bot commented Aug 17, 2026

Copy link
Copy Markdown

Preview deployment for your docs. Learn more about Mintlify Previews.

Project Status Preview Updated (UTC)
firecrawl 🟢 Ready View Preview Aug 17, 2026, 5:44 PM

💡 Tip: Enable Workflows to automatically generate PRs for you.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant