docs: add Plano, Repowise and PyGPT integration guides

[MVS-source]

What

The last three merged Eden AI integrations that had no page here. With #77 and #99, every merged integration is now documented.

Guide	Nav group	Upstream PR (merged)	Stars
plano.mdx	AI Gateway	katanemo/plano#1002, 2026-07-29	7.0k
repowise.mdx	AI Frameworks	repowise-dev/repowise#705, 2026-07-07	6.3k
py-gpt.mdx	Chat UI	szczyglis-dev/py-gpt#194, 2026-07-22	1.9k

Same format as the last integration PRs: frontmatter plus <TechArticleSchema>, docs.json nav registration, and a currentColor line icon in the style of the existing small ones. Each page lands last in its group, matching the add-order those groups already use.

Where the configuration comes from

Lifted from the merged upstream code, not reconstructed from memory.

Plano. The ProviderDefault in cli/planoai/defaults.py (env_var="EDENAI_API_KEY", base_url, model_pattern="edenai/*") and the wildcard config in their own cli/test/test_config_generator.py fixture, which is where the passthrough_auth: true form comes from. Two details from crates/hermesllm/src/providers/id.rs are worth having in the guide: ProviderId::EdenAI parses "edenai" and explicitly rejects eden_ai, and it carries no hardcoded model list, the same treatment as OpenRouter. It also maps both OpenAI and Anthropic client shapes to OpenAI chat completions upstream.

Repowise. The CLI flags and environment variables from the CONFIG.md and .env.example we contributed, including --reasoning and the fact that EDENAI_BASE_URL covers the LLM and the embedder in one setting. Eden AI is a provider on both sides here, which is unusual enough to be the main point of the page.

PyGPT. The api_key_edenai and api_endpoint_edenai settings keys, their default endpoint value, the Eden AI settings tab, and the fact that the provider declares both LlamaIndex mode and embeddings.

Verification

• Every model id cited was checked against the live catalogues. Eight ids, zero dead.
• All twelve internal links resolve to existing pages.
• The three SVGs parse as XML.
• docs.json parses.

Notes

The datePublished and dateModified values in the schema blocks are placeholders. update-dates.yml sets both for added .mdx files, so the bot will normalise them when this PR opens.

Two gotchas I put in the guides on purpose, because both have cost time before:

• The chat and embedding catalogues are separate endpoints. An embedding model does not appear in /v3/models, which is where people look first.
• A model's vector width is a property of the gateway serving it, not of the model name. The same identifier can return a different number of dimensions through two different gateways, so the Repowise page tells the reader to let it read the width from Eden AI rather than copy a figure from another provider's table.

Execution Tests will be red here as it is on every PR in this repo since around 11 August: it dies in pytest_sessionstart on tests/helpers/api.py getting 401 Unauthorized from https://staging-api.edenai.run/v2/user/custom_token/. Unrelated to this change, and worth fixing separately.

[mintlify]

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

Project	Status	Preview	Updated (UTC)
edenai	🟢 Ready	View Preview	Aug 31, 2026, 4:08 PM

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

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 1307641c-6c98-4a27-a570-4905dad365a1

---

_{Comment @coderabbitai help to get the list of available commands.}