[adityabatra072]

Related issue

Fixes #8784

Summary

Add a Wally accordion to the provider examples in concepts/llms.mdx. Wally serves open-weight models (GLM, DeepSeek, Kimi, MiniMax) behind an OpenAI-compatible API, so the example uses the existing openai/ prefix with base_url and goes through the native OpenAI client. No code change, no LiteLLM.

llm = LLM(
    model="openai/glm-5.3-flash",
    base_url="https://inference.runanywhere.ai/v1",
    api_key=os.getenv("WALLY_API_KEY"),
)

Docs accordion added in en, ar, ko and pt-BR - code blocks identical, only prose translated.

Verification

• Tests added or updated for the changed behavior

• Relevant tests and quality checks pass locally

• the docs snippet run against the live API on current main: plain call with every Wally model id; streaming, tool call through an Agent and response_format structured output with glm-5.3-flash and deepseek-v4.1-flash - all pass

• mint broken-links - no broken links in the changed pages

• git diff --check - clean

Additional context

The change was written with AI assistance (Claude Code), so per CONTRIBUTING the llm-generated label applies - I can't set labels on this repo, please add it.

---

Note

Low Risk
Documentation-only updates with no runtime or auth logic changes; risk is limited to doc accuracy and external link correctness.

Overview
Adds a Wally provider section to the LLM provider examples in concepts/llms.mdx, addressing #8784. The new accordion sits after DeepInfra and documents Wally’s OpenAI-compatible endpoint: set WALLY_API_KEY, point the native OpenAI client at https://inference.runanywhere.ai/v1 via base_url, and use models like openai/glm-5.3-flash (no LiteLLM extra). It also covers env setup, a sample LLM(...) block, feature bullets (open-weight models, tools/streaming/structured output, up to 1M context), and links to pricing and API docs.

The same accordion is added in en, ar, ko, and pt-BR under docs/edge/; code samples match across locales with translated prose only. No application code changes.

^{Reviewed by Cursor Bugbot for commit a9aa533. Bugbot is set up for automated code reviews on this repo. Configure here.}

[mintlify]

Preview approval required

This pull request comes from a fork, so Mintlify won't build a preview until someone with write access approves commit a9aa533.

Approve preview

[strix-security]

Strix Security Review

Warning

This pull request has 9 commits after the last Strix review (607ac97). Strix has not reviewed these changes.
Automatic review on push is off for this repository. To review the latest changes, tag @strix-security in a comment, or turn on re-review on push.

✅ No open security findings.

PR summary: Adds a Wally provider example to the LLM provider accordion in the English, Arabic, Korean and Brazilian Portuguese editions of `concepts/llms.mdx`, showing how to reach Wally's OpenAI-compatible endpoint through the native OpenAI client with a custom `base_url` and an API key read from an environment variable.

Review summary

This PR is documentation-only: it adds a Wally provider example as a Mintlify accordion to concepts/llms.mdx in the en, ar, ko and pt-BR locales. All four hunks are prose plus illustrative snippets; there is no production code, runtime config, build/CI change, or dependency change in the diff, so there is no changed code path an attacker could reach. The snippets use a placeholder for the key in the .env sample and read it via os.getenv("WALLY_API_KEY") rather than embedding a credential, and the custom base_url pattern matches the two dozen other provider examples already present in the same file. No security issues found.

_{Review 1 on this PR · 607ac97 · requested Oct 29 19:42 UTC when the PR was opened}

---

Reviewed by Strix
_{Re-run review · Configure security review settings}