Upgrade LiteLLM, add new models

[bubacoder]

Summary by CodeRabbit

• Chores

  • Updated AI service container image to latest version (v1.80.0-stable)

• New Features

  • Added container health monitoring with automatic restart capability

• Updates

  • Updated Claude model to latest version
  • Added support for Grok model

_{✏️ Tip: You can customize this high-level summary in your review settings.}

[coderabbitai]

Walkthrough

This PR updates the LiteLLM Docker configuration by upgrading the container image to v1.80.0-stable, adds healthcheck configuration, modifies volume mounting strategy, and updates model configurations to use Claude Opus 4.5 and add OpenRouter Grok support.

Changes

Cohort / File(s)	Summary
Docker container configuration docker/ai/litellm.yaml	Upgraded container image from v1.53.1 to v1.80.0-stable, converted command to array format with flags (--config=/app/config.yaml, --telemetry=false), changed volume mount from directory to specific config file, and added healthcheck with 30s interval and 40s start period.
LiteLLM model configuration docker/ai/litellm/config.yaml	Updated Anthropic model entry from claude-opus-4-1 to claude-opus-4-5, added new OpenRouter model entry for grok-code-fast-1.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

• Straightforward version and configuration updates
• Healthcheck configuration is standard Docker syntax
• Model configuration changes are localized with no cross-file dependencies

Possibly related PRs

• Update Anthropic models #213: Updates Anthropic model entries in docker/ai/litellm/config.yaml that are further refined by this PR's claude-opus-4-5 upgrade.
• Add new AI models, update docs #137: Adds LiteLLM model entries (Anthropic/OpenRouter providers) in the same configuration file, establishing precedent for the model configuration pattern.

Pre-merge checks and finishing touches

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the main changes: upgrading LiteLLM from v1.53.1 to v1.80.0 and adding a new Grok model.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch feature/litellm

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[coderabbitai]

Actionable comments posted: 1

📜 Review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 2c1af56 and db1b27b.

📒 Files selected for processing (2)

• docker/ai/litellm.yaml (2 hunks)
• docker/ai/litellm/config.yaml (2 hunks)

🧰 Additional context used

📓 Path-based instructions (1)

**/*.{yaml,yml}

📄 CodeRabbit inference engine (AGENTS.md)

YAML files must pass YAML linting and validation

Files:

• docker/ai/litellm.yaml
• docker/ai/litellm/config.yaml

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)

• GitHub Check: check

🔇 Additional comments (5)

docker/ai/litellm.yaml (5)

33-36: ✓ Claude Opus 4.5 model reference is valid.

The model identifier claude-opus-4-5 is the correct way to access Claude Opus 4.5 via the Claude API, and the configuration properly references it as anthropic/claude-opus-4-5.

---

117-122: ✓ Grok Code Fast 1 model reference is valid.

The model identifier x-ai/grok-code-fast-1 is the correct OpenRouter model path for accessing Grok Code Fast 1, and the configuration properly references it as openrouter/x-ai/grok-code-fast-1.

---

29-34: The healthcheck configuration is correct. LiteLLM v1.80.0 exposes the /health/liveliness endpoint as a liveness probe for container restarts, and it is unprotected by default, making it suitable for the wget-based healthcheck. The configuration follows documented best practices.

---

19-19: This volume mount configuration is correct as written.

The ./litellm directory contains only config.yaml, making the file-specific mount appropriate. Since LiteLLM v1.80.0 is invoked with --config=/app/config.yaml and no other files are present in the source directory, the file-only mount is the correct approach. YAML validation passes.

Likely an incorrect or invalid review comment.

---

13-15: Command format is compatible with LiteLLM v1.80.0.

The --config=/app/config.yaml and --telemetry=false flags are both supported by LiteLLM v1.80.0 proxy CLI. The YAML array format is correctly structured for the container entrypoint. No compatibility issues found.