[Feat] Add Claude Opus 5.5 to model providers · Pull Request #1756 · Zoo-Code-Org/Zoo-Code

ghost · 2026-09-22T22:46:12Z

Opened on behalf of @taltas. Follow up by mentioning @roomote, in the web UI, or in Discord.

Related GitHub Issue

No tracking issue — this model addition was requested directly (Anthropic released Claude Opus 5.5 on Sept 22, 2026: https://www.anthropic.com/claude-opus-5-5).

Description

Adds Anthropic's newly released Claude Opus 5.5 (API model id claude-opus-5-5) to every provider surface where claude-opus-5 exists, mirroring its configuration exactly:

Model registries (packages/types/src/providers/): new entries in anthropic.ts, vertex.ts, bedrock.ts (including BEDROCK_GLOBAL_INFERENCE_MODEL_IDS and its availability comment), openrouter.ts (prompt-caching and reasoning-budget sets), and vercel-ai-gateway.ts (prompt-caching and vision-and-tools sets), using each provider's naming convention (anthropic.claude-opus-5-5 on Bedrock, anthropic/claude-opus-5-5 on OpenRouter/Vercel AI Gateway).
Anthropic handler (src/api/providers/anthropic.ts): added claude-opus-5-5 to both switch statements (prompt-caching request path and prompt-caching-2024-07-31 beta header).
Model fetchers (src/api/providers/fetchers/): per-model overrides in openrouter.ts (Anthropic maxTokens, binary reasoning, no temperature), requesty.ts (reasoning budget/binary, no temperature), and vercel-ai-gateway.ts (no temperature). Bedrock's isAdaptiveThinkingModel already matches opus-5-5 via its existing includes("opus-5") pattern guard.
Configuration mirrors Opus 5: 1M native context window, reasoning budget + binary toggle, supportsTemperature: false, images and prompt caching supported. Max output tokens: 128k on the direct Anthropic path and on Bedrock (per the AWS Bedrock model card, which also sets the Bedrock cache-checkpoint minimum to 512, matching the Fable 5.1 entry); 8192 on Vertex (same as Opus 5).
Pricing per 1M tokens from the announcement: input $4, output $20, cache write $5, cache read $0.20.

Test Procedure

Added mirrored test coverage everywhere Opus 5 had dedicated tests: anthropic.spec.ts (model info + adaptive-thinking request with prompt-caching beta), anthropic-vertex.spec.ts (model info + adaptive thinking), bedrock.spec.ts (model info, global-inference prefix, adaptive-thinking predicate incl. global. prefix), and the openrouter/requesty/vercel-ai-gateway fetcher specs (per-model overrides).
Targeted run: 9 spec files, 353 tests passed.
Full pnpm test from the repo root: 13/13 turbo tasks successful. pnpm check-types: 11/11 successful. Full pnpm lint (via pre-commit hook): 11/11 successful; ESLint suppression counts unchanged.
Reviewers can verify by selecting an Anthropic-compatible provider and confirming claude-opus-5-5 resolves with 1M context, $4/$20 pricing, and adaptive-thinking behavior (no temperature sent).

Pre-Submission Checklist

Issue Linked: No approved issue exists; this is a direct model-addition request (see "Related GitHub Issue").
Scope: My changes are focused on the linked issue (one major feature/fix per PR).
Self-Review: I have performed a thorough self-review of my code.
Testing: New and/or updated tests have been added to cover my changes (if applicable).
Visual Snapshot (UI changes only): Not applicable — no UI change; the new model appears automatically in provider model pickers from the registry entries.
Documentation Impact: I have considered if my changes require documentation updates (see "Documentation Updates" section below).
Contribution Guidelines: I have read and agree to the Contributor Guidelines.

Visual Snapshots

Not applicable — no user-visible layout, theme, or rendered-state change.

Videos (interaction / animation only)

Not applicable.

Documentation Updates

No documentation updates are required.

Checked the Zoo-Code-Docs repository: provider docs do not enumerate specific Claude model versions (only generic setup instructions), so there is no model list to update. Historical update-notes were left untouched.

Additional Notes

The 1M context window and 128K max output tokens are confirmed by the AWS Bedrock model card: https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-anthropic-claude-opus-5-5.html. Optional fast-mode pricing ($8/$40) is not modeled since Opus 5 has no equivalent tier configuration in the codebase.

Get in Touch

Discord: taltas1

github-actions · 2026-09-22T22:47:24Z

Review status

This PR was opened by an automated account. A human maintainer must verify the change intent, provenance, and validation before merging.

Current step: Required CI passed. Waiting for automated review of the latest commit.

If automated review does not start, a maintainer must restart it.

Review-state labels are managed by this workflow; do not edit them manually.

coderabbitai · 2026-09-22T22:47:30Z

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Repository: Zoo-Code-Org/Zoo-Code/.coderabbit.yaml

Review profile: ASSERTIVE

Plan: Advanced

Run ID: 52d93d3f-f4dc-40a3-bdad-714f5cd8d967

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

🔍 Trigger review

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.}

codecov · 2026-09-22T22:57:32Z

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

edelauna · 2026-09-23T01:21:14Z

@coderabbitai review

edelauna · 2026-09-23T04:12:50Z

+	if (id === "anthropic/claude-opus-5-5") {
+		modelInfo.supportsTemperature = false
+	}


Shouldn't this block also set supportsReasoningBudget = true and supportsReasoningBinary = true? Every other provider surface for claude-opus-5-5 (Bedrock, Vertex, direct Anthropic, OpenRouter, Requesty) sets both flags, but no Anthropic model in this fetcher gets them — OpenAI models in the same file do. Is the omission intentional for the Vercel AI Gateway path?

Vercel lists Opus 5.5 as supporting effort-based reasoning, not a fixed budget or an on/off toggle, so these flags do not apply: docs, model page.

edelauna · 2026-09-23T04:12:51Z

+				model: mockModel,
+				inputModality: ["text", "image"],
+				outputModality: ["text"],
+				maxTokens: 128000,


The fixture passes maxTokens: 128000, which matches anthropicModels["claude-opus-5-5"].maxTokens exactly — so deleting the override at openrouter.ts:314 would still pass this assertion. Would using a different fixture value (e.g. 16384) with the same toBe(128000) expectation prove the override actually fires? The claude-sonnet-4.6 test uses this pattern.

The test now uses a different input value to verify the override.

… minimum per AWS model card

nh2 · 2026-09-24T11:01:05Z

+	"anthropic.claude-opus-5-5": {
+		// 128K max output tokens per the AWS Bedrock model card:
+		// https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-anthropic-claude-opus-5-5.html
+		maxTokens: 128_000,


What about the supportsMaxTokens: true, from

https://github.com/Zoo-Code-Org/Zoo-Code/pull/1718/changes#diff-e5f2e068d695776d0c7b28f26574ccb565232f0db7f03882a4260eed155eb6ccR241

Is that needed/correct?

What about supportsReasoningBinary?

Changes made

nh2 · 2026-09-26T15:34:45Z

@taltas Did you or @zoomote read the Opus 5.5 migration guide?

https://platform.claude.com/docs/en/models/opus-5-5/migration-guide#text-between-tool-calls

Or this? https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-opus-5-5#user-facing-progress-updates

If not, I suppose this may cause:

[BUG] Cannot make Opus 5.5 describe commands before running them #1820

feat(providers): add Anthropic Claude Opus 5.5 (claude-opus-5-5)

2557117

ghost assigned taltas Sep 22, 2026

taltas marked this pull request as ready for review September 22, 2026 22:49

taltas requested review from JamesRobert20, edelauna, hannesrudolph, navedmerchant and taltas as code owners September 22, 2026 22:49

github-actions Bot added coderabbit-review-active Required CI passed; CodeRabbit review is active awaiting-coderabbit Waiting for CodeRabbit to approve the latest commit labels Sep 22, 2026

edelauna requested changes Sep 23, 2026

View reviewed changes

github-actions Bot added awaiting-author PR is waiting for the author to address requested changes and removed coderabbit-review-active Required CI passed; CodeRabbit review is active awaiting-coderabbit Waiting for CodeRabbit to approve the latest commit labels Sep 23, 2026

taltas requested a review from edelauna September 23, 2026 14:27

nh2 suggested changes Sep 23, 2026

View reviewed changes

Comment thread packages/types/src/providers/bedrock.ts Resolved

fix(bedrock): correct Opus 5.5 max output tokens and cache checkpoint…

666a567

… minimum per AWS model card

github-actions Bot added awaiting-author PR is waiting for the author to address requested changes and removed awaiting-author PR is waiting for the author to address requested changes labels Sep 24, 2026

edelauna approved these changes Sep 24, 2026

View reviewed changes

github-actions Bot added coderabbit-review-active Required CI passed; CodeRabbit review is active awaiting-coderabbit Waiting for CodeRabbit to approve the latest commit and removed awaiting-author PR is waiting for the author to address requested changes labels Sep 24, 2026

nh2 previously requested changes Sep 24, 2026

View reviewed changes

taltas approved these changes Sep 24, 2026

View reviewed changes

Merge branch 'main' into add-claude-opus-5-5

61b4300

taltas added this pull request to the merge queue Sep 24, 2026

github-merge-queue Bot removed this pull request from the merge queue due to failed status checks Sep 24, 2026

taltas added this pull request to the merge queue Sep 24, 2026

Merged via the queue into main with commit 9ec139c Sep 24, 2026
16 of 17 checks passed

taltas deleted the add-claude-opus-5-5 branch September 24, 2026 13:58

wrsfmss mentioned this pull request Sep 25, 2026

fix(vertex): correct Claude Opus 5.5 max output tokens to 128K #1777

Merged

2 tasks done

This was referenced Sep 26, 2026

Add support for Bedrock to run OPUS 5.5 #1759

Open

[BUG] Cannot make Opus 5.5 describe commands before running them #1820

Open

ghost mentioned this pull request Sep 28, 2026

[Feat] Add Claude Sonnet 5.5 to model providers #1849

Closed

8 tasks done

Navigation Menu

Repository navigation

[Feat] Add Claude Opus 5.5 to model providers - #1756

[Feat] Add Claude Opus 5.5 to model providers#1756
taltas merged 3 commits into
mainfrom
add-claude-opus-5-5

ghost commented Sep 22, 2026 •

edited by ghost

Loading

Uh oh!

github-actions Bot commented Sep 22, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot commented Sep 22, 2026 •

edited

Loading

Review skipped

Uh oh!

codecov Bot commented Sep 22, 2026

Uh oh!

edelauna commented Sep 23, 2026

Uh oh!

edelauna Sep 23, 2026

Uh oh!

taltas Sep 23, 2026

Uh oh!

edelauna Sep 23, 2026

Uh oh!

taltas Sep 23, 2026

Uh oh!

Uh oh!

nh2 Sep 24, 2026

Uh oh!

Uh oh!

Uh oh!

nh2 commented Sep 26, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

ghost commented Sep 22, 2026 • edited by ghost Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Related GitHub Issue

Description

Test Procedure

Pre-Submission Checklist

Visual Snapshots

Videos (interaction / animation only)

Documentation Updates

Additional Notes

Get in Touch

Uh oh!

github-actions Bot commented Sep 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review status

Uh oh!

coderabbitai Bot commented Sep 22, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Review skipped

Uh oh!

codecov Bot commented Sep 22, 2026

Codecov Report

Uh oh!

edelauna commented Sep 23, 2026

Uh oh!

edelauna Sep 23, 2026

Choose a reason for hiding this comment

Uh oh!

taltas Sep 23, 2026

Choose a reason for hiding this comment

Uh oh!

edelauna Sep 23, 2026

Choose a reason for hiding this comment

Uh oh!

taltas Sep 23, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

nh2 Sep 24, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

nh2 commented Sep 26, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

ghost commented Sep 22, 2026 •

edited by ghost

Loading

github-actions Bot commented Sep 22, 2026 •

edited

Loading

coderabbitai Bot commented Sep 22, 2026 •

edited

Loading