feat(models): add deepseek-v4-flash-0731

[smakosh]

Summary

DeepSeek released DeepSeek V4 Flash (0731) today (Jul 31, 2026) — a dated snapshot of V4 Flash. This PR adds it to the catalogue and announces it on the public changelog.

Model (packages/models)

• New deepseek-v4-flash-0731 entry with a first-party deepseek provider mapping:
  • $0.14/M input, $0.28/M output, $0.0028/M cached input
  • 1,050,000-token context, 393,216 max output
  • Reasoning (none/high/max efforts), tools, JSON output, streaming — mirrors the undated deepseek-v4-flash deepseek mapping

Changelog (apps/ui)

• New entry 2026-07-31-deepseek-v4-flash-0731 (id 73) with pricing table, capabilities, and curl example
• OG image generated with gpt-image-2 in the house circuit-board style, featuring the DeepSeek whale on the central chip

Validation

• pnpm build — all 17 tasks pass (ui build validates the changelog frontmatter schema)
• packages/models unit tests — 108 passed
• pnpm format — clean

https://claude.ai/code/session_019HckdHtawypLYPeNF53NtA

Summary by CodeRabbit

• New Features
  • Added DeepSeek V4 Flash (0731) AI model featuring advanced reasoning capabilities, tool calling, structured outputs, and streaming support. The model includes a 1.05M token context window and 393K token maximum output limit, with competitive pricing and prompt/response caching. Available on all pricing tiers.

[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 Plus

Run ID: 257d1b0f-d22b-4713-84c7-a29cdf993dd5

📥 Commits

Reviewing files that changed from the base of the PR and between 8510372 and 9425cd5.

⛔ Files ignored due to path filters (1)

• apps/ui/public/changelog/deepseek-v4-flash-0731.png is excluded by !**/*.png

📒 Files selected for processing (2)

• apps/ui/src/content/changelog/2026-07-31-deepseek-v4-flash-0731.md
• packages/models/src/models/deepseek.ts

---

Walkthrough

Adds DeepSeek V4 Flash (0731) to the model catalog and documents its pricing, limits, capabilities, API usage, dated-model behavior, and availability.

Changes

DeepSeek V4 Flash

Layer / File(s)	Summary
Model metadata and documentation packages/models/src/models/deepseek.ts, apps/ui/src/content/changelog/...	Adds the model configuration with pricing, token limits, reasoning, streaming, vision, tools, JSON, and request parameter support. Adds a changelog page with usage and availability details.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related PRs

• theopenco/llmgateway#3072: Updates reasoning configuration for DeepSeek V4 models.
• theopenco/llmgateway#3131: Configures reasoning capabilities in DeepSeek model definitions.
• theopenco/llmgateway#3159: Updates supported parameters for DeepSeek V4 models.

Suggested reviewers: steebchen, amineace

🚥 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 identifies the primary change: adding the dated DeepSeek V4 Flash model definition.
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.

✨ 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 feat/deepseek-v4-flash-0731

---

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.}

[steebchen]

Images automagically compressed by Calibre's image-actions ✨

Compression reduced images by 71.6%, saving 1.5 MB.

Filename	Before	After	Improvement	Visual comparison
apps/ui/public/changelog/deepseek-v4-flash-0731.png	2.1 MB	597.7 KB	71.6%	View diff