β¨ feat(model-bank): add claude-fable-5 to Anthropic models#15639
Conversation
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
There was a problem hiding this comment.
π‘ Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: b271fae3c0
βΉοΈ 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 π.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| { name: 'textInput_cacheRead', rate: 1, strategy: 'fixed', unit: 'millionTokens' }, | ||
| { name: 'textInput', rate: 10, strategy: 'fixed', unit: 'millionTokens' }, | ||
| { name: 'textOutput', rate: 50, strategy: 'fixed', unit: 'millionTokens' }, | ||
| { name: 'textInput_cacheWrite', rate: 12.5, strategy: 'fixed', unit: 'millionTokens' }, |
There was a problem hiding this comment.
Price 1-hour cache writes correctly
For Claude Fable 5, Anthropic's current pricing table lists prompt-cache writes at $12.50/MTok for 5 minutes and $20/MTok for 1 hour, but this fixed textInput_cacheWrite rate only represents the 5-minute case. In contexts where the cost UI/calculator is asked to price a 1-hour cache TTL, this new model will under-report cache-write cost instead of using the existing lookup shape used by other Anthropic models with ttl pricing.
Useful? React with πΒ / π.
| releasedAt: '2026-06-09', | ||
| settings: { | ||
| disabledParams: ['temperature', 'top_p'], | ||
| extendParams: ['disableContextCaching', 'enableAdaptiveThinking', 'opus47Effort'], |
There was a problem hiding this comment.
Remove the fake adaptive-thinking toggle
Claude Fable 5 has adaptive thinking always on; the Anthropic docs say omitting thinking still runs adaptive thinking and thinking: {type: 'disabled'} is unsupported. Including enableAdaptiveThinking here exposes the existing on/off control, but when users turn it off modelParamsResolver merely omits the field, so the setting cannot disable anything for this model and the UI presents a no-op control as if it changed behavior.
Useful? React with πΒ / π.
β¦ a bound device Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Codecov Reportβ
All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## canary #15639 +/- ##
=======================================
Coverage 67.65% 67.65%
=======================================
Files 3360 3360
Lines 339184 339215 +31
Branches 36610 36613 +3
=======================================
+ Hits 229462 229503 +41
+ Misses 109531 109521 -10
Partials 191 191
Flags with carried forward coverage won't be shown. Click here to find out more.
π New features to boost your workflow:
|
# π LobeHub Release (20260610) **Release Date:** June 10, 2026 **Since v2.2.2:** 131 merged PRs Β· 13 contributors > This weekly release strengthens agent collaboration across cloud, desktop, CLI, and workspace flows, with steadier runtime behavior and a broader foundation for workspace-scoped data. --- ## β¨ Highlights - **Agent execution across devices** β Unifies per-device working directories, project skill discovery, and sub-agent suspend/resume behavior across server, QStash, and device RPC flows. (#15543, #15566, #15481, #15620, #15591) - **Connector and sandbox platform** β Expands connector permissions, custom OAuth MCP connector onboarding, sandbox provider support, and user-uploaded file sync into cloud sandbox runs. (#15463, #15546, #15184, #15550) - **Desktop and CLI reliability** β Fixes desktop cold-start, auto-update, Windows build, CLI skill discovery, and `lh connect` agent dispatch paths. (#15547, #15525, #15527, #15562, #15632, #15634) - **Pages and sharing** β Refreshes topic sharing, improves Page Editor layout behavior, and routes Page Agent tool execution through the server-side editor path. (#15581, #15556, #15588, #15023, #15610) - **Model availability and provider updates** β Adds user-scoped LobeHub model availability, Claude Fable 5, Qwen thinking preservation, and MiniMax M3 updates. (#15590, #15639, #13494, #15376) --- ## ποΈ Core Product & Architecture ### Agent Runtime & Heterogeneous Agents - Improves sub-agent lifecycle handling, including async suspend/resume, queue-mode QStash resume delivery, and blocking nested sub-agent calls. (#15481, #15620, #15575) - Stabilizes heterogeneous agent ingestion and streaming with raw stream dumps, per-turn usage, image forwarding on regenerate, and duplicate-text fixes. (#15602, #15577, #15592, #15585) - Adds execution-device and working-directory controls across device RPC, legacy defaults, and remote-spawned Claude Code sessions. (#15543, #15566, #15591, #15572) - Improves runtime diagnostics and compatibility, including Gemini multimodal output capture, abort stream semantics, and trace quality analysis. (#15535, #13677, #15508) --- ## π± Platforms, Integrations & UX ### Connectors, Sandbox & Tools - Ships API-level connector tool permissions, custom OAuth MCP connector onboarding, and connector-first runtime execution. (#15463, #15546) - Adds sandbox provider support, cloud sandbox file sync, and safer external URL file input handling with SSRF validation. (#15184, #15550, #12657) - Improves tool visibility and execution with pinned app-fixed tools, ANSI output rendering, gateway-tunneled MCP calls, and automatic headless tool runs. (#15509, #15516, #15469, #15492) ### Desktop, CLI & Web UX - Restores desktop startup and reload behavior, preserves IPC error causes, and keeps the tab bar new-tab action visible across routes. (#15547, #15597, #15638) - Fixes desktop update and build stability for browser quit guards, macOS update signing, and Windows Visual Studio detection. (#15525, #15527, #15562) - Shows the plan-limit upgrade UI on desktop builds. (#15628) - Adds the Agent Run delivery checker and fixes CLI device dispatch plus skill list/search output. (#15489, #15634, #15632) - Refreshes onboarding, auth source preservation, topic UI states, referral/Fable campaign copy, and chat-input control bar behavior. (#15629, #15544, #15573, #15614, #15616, #15617, #15622, #15643) --- ## π Security, Reliability & Rollout Notes - External URL file input now includes SSRF validation for safer Google file handling. (#12657) - Database workspace-scope migrations are part of this release; self-hosted operators should run the normal migration path before serving the updated app. (#15446, #15465, #15468, #15472) - The release branch was re-cut from `canary` and includes the latest `main` release-version commit so `v2.2.2` is the verified compare base. --- ## π₯ Contributors @ONLY-yours, @sxjeru, @hardy-one, @xujingli, @hezhijie0327, @Coooolfan, @arvinxx, @tjx666, @Innei, @rivertwilight, @rdmclin2, @cy948, @AmAzing129 **Full Changelog**: v2.2.2...release/weekly-20260610-recut-3
π» Change Type
π Related Issue
π Description of Change
Adds Claude Fable 5 (
claude-fable-5, GA 2026-06-09) to the Anthropic model bank, placed above Opus 4.8 and enabled by default.Specs per Anthropic's official models documentation:
settingsmirror Opus 4.8 (disabledParams: ['temperature', 'top_p'],extendParams: ['disableContextCaching', 'enableAdaptiveThinking', 'opus47Effort']) β Fable 5 shares the Opus 4.7/4.8 API surface (adaptive thinking only, sampling params removed)Runtime safety note: Fable 5 returns 400 on an explicit
thinking: {type: 'disabled'}(the param must be omitted instead). VerifiedanthropicCompatibleFactoryalready omits thethinkingfield entirely when thinking is not enabled/adaptive, so the existingenableAdaptiveThinkingtoggle is safe with this model and no runtime change is needed.π§ͺ How to Test
claude-fable-5: adaptive thinking toggle and effort selector work; temperature/top_p are hidden.cd packages/model-bank && bunx vitest run src/aiModels/__tests__/index.test.tspasses.πΈ Screenshots / Videos
π Additional Information
Out of scope (left unchanged on purpose): the home-page starter model constant (
starterModels.ts) still promotes Claude Opus 4.8 β switching the default recommendation to Fable 5 is a product decision.π€ Generated with Claude Code