feat(model):add wandb model offerings to include kimi-k2.5 and minimax-m2.5

[csoni-cweave]

Relevant issues

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/test_litellm/ directory, Adding at least 1 test is a hard requirement - see details
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem
• I have requested a Greptile review by commenting @greptileai and received a Confidence Score of at least 4/5 before requesting a maintainer review

Delays in PR merge?

If you're seeing a delay in your PR being merged, ping the LiteLLM Team on Slack (#pr-review).

CI (LiteLLM team)

CI status guideline:

• 50-55 passing tests: main is stable with minor issues.
• 45-49 passing tests: acceptable but needs attention
• <= 40 passing tests: unstable; be careful with your merges and assess the risk.

• Branch creation CI run
  Link:

• CI run for the last commit
  Link:

• Merge / cherry-pick CI run
  Links:

Type

🆕 New Feature

Changes

• add wandb model offerings to include kimi-k2.5 and minimax-m2.5

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
litellm	Ready	Preview, Comment	Apr 9, 2026 9:01am

[csoni-cweave]

@greptileai

[greptile-apps]

Greptile Summary

This PR adds two new WandB model offerings — wandb/moonshotai/Kimi-K2.5 and wandb/MiniMaxAI/MiniMax-M2.5 — to the model price/context window maps and the WANDB_MODELS set in constants.py. Both JSON files (main and backup) are kept in sync, and a new unit test verifies the pricing entries for both models.

Confidence Score: 5/5

Safe to merge — purely additive model metadata with no logic changes.

All four files are consistent: constants.py, both JSON files, and the test are in sync. The test uses a local cost map (no network calls), follows the same pattern as existing pricing tests, and correctly asserts the new pricing values. No existing tests are weakened or removed.

No files require special attention.

Vulnerabilities

No security concerns identified. Changes are purely additive model metadata (pricing, context windows, capability flags) with no auth, input handling, or network path changes.

Important Files Changed

Filename	Overview
litellm/constants.py	Adds moonshotai/Kimi-K2.5 and MiniMaxAI/MiniMax-M2.5 to the WANDB_MODELS set, consistent with existing entries and JSON keys.
model_prices_and_context_window.json	Adds wandb/moonshotai/Kimi-K2.5 and wandb/MiniMaxAI/MiniMax-M2.5 with pricing, context window, capability flags, and source URLs — matches the backup file.
litellm/model_prices_and_context_window_backup.json	Kept in sync with the main JSON file; includes both new WandB model entries.
tests/test_litellm/test_cost_calculator.py	Adds test_wandb_model_api_pricing_entries that uses a local cost map (no network calls) to assert input/output pricing for both new models.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[WANDB_MODELS set\nlitellm/constants.py] -->|includes| B["moonshotai/Kimi-K2.5"]
    A -->|includes| C["MiniMaxAI/MiniMax-M2.5"]
    B --> D["wandb/moonshotai/Kimi-K2.5\nin JSON price map\n262K ctx, $0.60/$3.00 per M tokens\nsupports reasoning, vision, function calling"]
    C --> E["wandb/MiniMaxAI/MiniMax-M2.5\nin JSON price map\n197K ctx, $0.30/$1.20 per M tokens\nsupports reasoning, function calling"]
    D --> F["model_prices_and_context_window.json\n+ backup.json"]
    E --> F
    F --> G["test_wandb_model_api_pricing_entries\nasserts input/output cost per token"]

Loading

_{Reviews (2): Last reviewed commit: "feat(model):add wandb model offerings to..." | Re-trigger Greptile}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[codspeed-hq]

Merging this PR will not alter performance

✅ 16 untouched benchmarks

---

_{Comparing csoni-cweave:add-wandb-models (249c308) with main (fdd7500)¹}

Footnotes

1. No successful run was found on litellm_oss_staging_04_11_2026 (97f722f) during the generation of this report, so main (fdd7500) was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩