feat(models): add Seedream 5.0 Pro and Lite by steebchen · Pull Request #2960 · theopenco/llmgateway

steebchen · 2026-07-09T11:55:20Z

Summary

Adds ByteDance Seedream 5.0 Pro and Seedream 5.0 Lite to the gateway's image-generation models (bytedance / BytePlus ModelArk provider), and corrects Pro sizing/pricing after live testing.

model id	externalId	price	sizes
`seedream-5-0-pro`	`dola-seedream-5-0-pro-260628`	`0.09` / image	1K, 2K
`seedream-5-0-lite`	`seedream-5-0-260128`	`0.035` / image	2K, 4K

Both follow the existing seedream-4-5 mapping shape (imageGenerations: true, output: ["text","image"], text-to-image only, test: "skip").

Research / pricing

Model ids confirmed against the live BytePlus catalog (GET /api/v3/models).

Lite — flat $0.035/image, no resolution tiers (BytePlus official; corroborated by Lumenfall, EvoLink, Atlas). Branded "Dola Seedream 5.0 lite"; catalog id seedream-5-0-260128, released 2026-01-28.
Pro — tiered: $0.045 (≤2.36M px) / $0.09 (>2.36M px, i.e. 2K), no 4K at launch. The playground only exposes 2K/4K for seedream, and 2K is the $0.09 tier, so the margin-safe flat requestPrice is 0.09 (corrected from the earlier 0.06 estimate). Reference images: first free, +$0.003 each (not billed separately here).

Live verification (against real, activated model ids)

Ran the gateway image e2e (TEST_IMAGE_MODE=true TEST_IMAGE_SIZE=2K) per model:

endpoint	Pro	Lite
`/v1/chat/completions`	✅	✅
`/v1/images/generations`	✅	✅
`/v1/images/edits`	400 (expected)	400 (expected)

/v1/images/edits fails for every seedream model (text-to-image only; the bytedance image builder forwards just {model, prompt, size}, so vision: false is correct and consistent — the existing seedream-4-5 fails this same test identically).

Direct BytePlus probes established the size support that drove the playground change:

Pro: 1K ✅, 2K ✅, 1.5K 400, 4K 400 → max ~2K, no 4K at launch
Lite: 2K ✅, 4K ✅

Playground

getModelImageConfig now special-cases Seedream 5.0 Pro to offer 1K/2K (was 2K/4K), since Pro's 4K preset returns 400. Lite keeps the standard seedream 2K/4K.
The existing model-change reset effect already snaps a stale 4K selection back to the 2K default when switching to Pro, so no invalid size can be sent.
Both models appear in the image model selector automatically (they have image in output).

Notes

Latency: Pro 2K generations take ~120–135s upstream (within timeouts, served fine). Lite is much faster (~30s).

🤖 Generated with Claude Code

Summary by CodeRabbit

New Features
- Added support for two new ByteDance Seedream 5.0 image generation models: Lite and Pro.
- Updated the playground to recognize the Pro variant and show the correct available image size options.
- Seedream 5.0 now exposes image generation capabilities with preset limits and supported output sizes.

Add ByteDance Seedream 5.0 Pro (dola-seedream-5-0-pro-260628) as a text-to-image model on the bytedance provider, following the existing seedream-4-5 mapping. Verified end-to-end through the gateway image pipeline (chat completions + /v1/images/generations). The playground already handles it via the seedream image config (2K/4K sizes). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Copilot

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

coderabbitai · 2026-07-09T11:55:53Z

Walkthrough

Adds Seedream 5.0 Lite and Pro entries to the ByteDance model registry and updates playground image-size selection so Seedream 5.0 Pro uses its own preset sizes.

Changes

ByteDance Seedream 5.0 update

Layer / File(s)	Summary
Model registry entries `packages/models/src/models/bytedance.ts`	Adds `seedream-5-0-lite` and `seedream-5-0-pro` with release metadata, pricing, fixed context/output limits, and image-generation-only flags.
Playground size selection `apps/playground/src/lib/image-gen.ts`	Detects `seedream-5-0-pro` and returns `["1K", "2K"]` before the broader Seedream size handling.

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

Possibly related PRs

theopenco/llmgateway#2270: Also extends packages/models/src/models/bytedance.ts with new ByteDance model entries in the shared registry.
theopenco/llmgateway#2864: Also updates apps/playground/src/lib/image-gen.ts size-tier logic for a specific model variant.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
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.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the main change: adding Seedream 5.0 Pro and Lite model support.

✨ 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/seedream-5-0-pro

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

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

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 0b381ddd4b

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-07-09T12:00:03Z

+			{
+				test: "skip",
+				providerId: "bytedance",
+				externalId: "dola-seedream-5-0-pro-260628",


Keep inactive Seedream Pro out of the catalog

This mapping is active immediately: test: "skip" only skips e2e selection, while /v1/models and resolveModelInfo() still expose and route bytedance/seedream-5-0-pro because there is no deactivatedAt. In the verified BytePlus account this exact external model returns ModelNotOpen, so any production request pinned to the new model will route to an upstream 502 until the Ark Console activation is completed; keep the mapping deactivated or gate the change on activation.

Useful? React with 👍 / 👎.

Add ByteDance Seedream 5.0 Lite (seedream-5-0-260128) at the official $0.035/image flat rate. Verified end-to-end through the gateway image pipeline (chat completions + /v1/images/generations). Correct Seedream 5.0 Pro after live testing against BytePlus: the Pro tops out at ~2K (the "4K" preset returns 400) and its 2K tier bills at $0.09/image, so bump requestPrice 0.06 -> 0.09, drop the 4K claim from its description, and restrict its playground sizes to 1K/2K. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

chatgpt-codex-connector · 2026-07-09T13:13:17Z

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.

Copilot AI review requested due to automatic review settings July 9, 2026 11:55

Copilot AI reviewed Jul 9, 2026

chatgpt-codex-connector Bot reviewed Jul 9, 2026

View reviewed changes

steebchen changed the title ~~feat(models): add Seedream 5.0 Pro~~ feat(models): add Seedream 5.0 Pro and Lite Jul 9, 2026

steebchen merged commit a107b6e into main Jul 9, 2026
18 checks passed

steebchen deleted the feat/seedream-5-0-pro branch July 9, 2026 13:28

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat(models): add Seedream 5.0 Pro and Lite#2960

feat(models): add Seedream 5.0 Pro and Lite#2960
steebchen merged 2 commits into
mainfrom
feat/seedream-5-0-pro

steebchen commented Jul 9, 2026 •

edited by coderabbitai Bot

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

coderabbitai Bot commented Jul 9, 2026 •

edited

Loading

❌ Failed checks (1 warning)

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

chatgpt-codex-connector Bot Jul 9, 2026

Uh oh!

chatgpt-codex-connector Bot commented Jul 9, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

steebchen commented Jul 9, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Research / pricing

Live verification (against real, activated model ids)

Playground

Notes

Summary by CodeRabbit

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Uh oh!

coderabbitai Bot commented Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

❌ Failed checks (1 warning)

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Jul 9, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot commented Jul 9, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

steebchen commented Jul 9, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Jul 9, 2026 •

edited

Loading