chore(deepseek): refresh V4 Pro and V4 Flash prices

[steebchen]

Summary

• Update DeepSeek V4 Pro pricing on the deepseek provider: input $1.74→$0.435, output $3.48→$0.87, cached $0.145→$0.003625 per M tokens
• Correct DeepSeek V4 Flash cached input price on the deepseek provider: $0.028→$0.0028 per M tokens

Test plan

• Verify pricing matches DeepSeek's published rates

🤖 Generated with Claude Code

Summary by CodeRabbit

• Chores
  • Updated pricing rates for DeepSeek v4-pro and v4-flash models

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Repository UI

Review profile: CHILL

Plan: Pro

Run ID: 95b5e694-7438-4c61-a0a1-739cf0ad7ba5

📥 Commits

Reviewing files that changed from the base of the PR and between dbc6d47 and 70c08c1.

📒 Files selected for processing (1)

• packages/models/src/models/deepseek.ts

---

Walkthrough

Updated pricing constants for two DeepSeek model entries: reduced inputPrice, outputPrice, and cachedInputPrice for deepseek-v4-pro; reduced cachedInputPrice for deepseek-v4-flash. All numeric values in model configuration objects were modified.

Changes

Cohort / File(s)	Summary
DeepSeek Pricing Updates packages/models/src/models/deepseek.ts	Updated per-token pricing constants for deepseek-v4-pro (input, output, and cached input prices) and deepseek-v4-flash (cached input price) to lower rates.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~3 minutes

Possibly related PRs

• feat(models): add deepseek v4 pro and flash #2073: Adjusts per-token pricing values for the same deepseek-v4-pro and deepseek-v4-flash model entries with directly overlapping changes.
• feat(together-ai): add GLM-5.1, Kimi K2.6, DeepSeek V4 Pro #2094: Updates deepseek-v4-pro model pricing configuration, sharing the same model entry affected in this PR.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: updating DeepSeek V4 Pro and V4 Flash pricing constants, which directly matches the changeset modifications.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch deepseek-prices

---

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.}

[Copilot]

Pull request overview

Updates the DeepSeek provider’s token pricing constants to reflect refreshed published rates for DeepSeek V4 Pro and a corrected cached-input rate for DeepSeek V4 Flash.

Changes:

• Update deepseek-v4-pro input/output/cached-input prices for the deepseek provider.
• Correct deepseek-v4-flash cached-input price for the deepseek provider.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.