This commit addresses "Max retries exceeded" errors during stress testing where
all accounts would become exhausted simultaneously due to short per-second rate
limits triggering cascading failures.
## Rate Limit Parser (`rate-limit-parser.js`)
- Remove 2s buffer enforcement that caused cascading failures when API returned
short reset times (200-600ms). Now adds 200ms buffer for sub-500ms resets
- Add `parseRateLimitReason()` for smart backoff based on error type:
QUOTA_EXHAUSTED, RATE_LIMIT_EXCEEDED, MODEL_CAPACITY_EXHAUSTED, SERVER_ERROR
## Message/Streaming Handlers
- Add per-account+model rate limit state tracking with exponential backoff
- For short rate limits (< 1 second), wait and retry on same account instead
of switching - prevents thundering herd when all accounts hit per-second limits
- Add throttle wait support for fallback modes (emergency/lastResort)
- Add `calculateSmartBackoff()` with progressive tiers by error type
## HybridStrategy (`hybrid-strategy.js`)
- Refactor `#getCandidates()` to return 4 fallback levels:
- `normal`: All filters pass (health, tokens, quota)
- `quota`: Bypass critical quota check
- `emergency`: Bypass health check when ALL accounts unhealthy
- `lastResort`: Bypass BOTH health AND token bucket checks
- Add throttle wait times: 500ms for lastResort, 250ms for emergency
- Fix LRU calculation to use seconds (matches opencode-antigravity-auth)
## Health Tracker
- Increase `recoveryPerHour` from 2 to 10 for faster recovery (1 hour vs 5 hours)
## Account Manager
- Add consecutive failure tracking: `getConsecutiveFailures()`,
`incrementConsecutiveFailures()`, `resetConsecutiveFailures()`
- Add cooldown mechanism separate from rate limits with `CooldownReason`
- Reset consecutive failures on successful request
## Base Strategy
- Add `isAccountCoolingDown()` check in `isAccountUsable()`
## Constants
- Replace fixed `CAPACITY_RETRY_DELAY_MS` with progressive `CAPACITY_BACKOFF_TIERS_MS`
- Add `BACKOFF_BY_ERROR_TYPE` for smart backoff
- Add `QUOTA_EXHAUSTED_BACKOFF_TIERS_MS` for progressive quota backoff
- Add `MIN_BACKOFF_MS` floor to prevent "Available in 0s" loops
- Increase `MAX_CAPACITY_RETRIES` from 3 to 5
- Reduce `RATE_LIMIT_DEDUP_WINDOW_MS` from 5s to 2s
## Frontend
- Remove `capacityRetryDelayMs` config (replaced by progressive tiers)
- Update default `maxCapacityRetries` display from 3 to 5
## Testing
- Add `tests/stress-test.cjs` for concurrent request stress testing
Co-Authored-By: Claude <noreply@anthropic.com>
* feat: add manual OAuth flow support in WebUI
* fix: reset add account modal state on close
* feat: display custom API key in startup banner
* fix: move translations to separate files and optimize import API
* fix: remove orphaned model-manager.js and cleanup callback server on manual auth
---------
Co-authored-by: Badri Narayanan S <59133612+badrisnarayanan@users.noreply.github.com>
When all accounts are rate-limited or token-exhausted, the retry loop
was incorrectly counting the wait time as a failed attempt. This caused
premature "Max retries exceeded" errors when we were just patiently
waiting for accounts to become available.
- Add attempt-- after sleeping for rate limits or strategy waits
- Add #diagnoseNoCandidates() to hybrid strategy for better logging
- Add getTimeUntilNextToken() and getMinTimeUntilToken() to token tracker
- Return waitMs from hybrid strategy when all accounts are token-blocked
Co-Authored-By: Claude <noreply@anthropic.com>
The POST /api/config endpoint was missing validation for several
Advanced Server Settings fields, causing the sliders to fail silently.
Added support for: rateLimitDedupWindowMs, maxConsecutiveFailures,
extendedCooldownMs, capacityRetryDelayMs, maxCapacityRetries.
Fixes#181
Co-Authored-By: Claude <noreply@anthropic.com>
The hybrid strategy now considers account quota levels when selecting
accounts, preventing any single account from being drained to 0%.
- Add QuotaTracker class to track per-account quota levels
- Exclude accounts with critical quota (<5%) from selection
- Add quota component to scoring formula (weight: 3)
- Fall back to critical accounts when no alternatives exist
- Add 18 new tests for quota-aware selection
Scoring formula: Health×2 + Tokens×5 + Quota×3 + LRU×0.1
An attempt at resolving badrisnarayanan/antigravity-claude-proxy#171
Adds `maxAccounts` configuration parameter to control the maximum number of Google accounts.
**Changes:**
- New config field `maxAccounts` (default: 10, range: 1-100)
- Settings page: slider control for adjusting limit
- Accounts page: counter badge (e.g., "8/10") with visual feedback
- Add button disabled when limit reached
- Server-side validation on account creation
**Breaking Changes:** None
Adds logging when 403/404 errors trigger endpoint fallback (daily → prod).
The retry behavior was already working but silently - now it's visible.
Matches opencode-antigravity-auth error handling behavior.
Co-Authored-By: Claude <noreply@anthropic.com>
Fixes regression where /v1/models returned 500 error because
pickNext() method was removed in v2.2.x refactor.
Fixes#164
Co-Authored-By: Claude <noreply@anthropic.com>
paidTier values like g1-pro-tier and g1-ultra-tier are rejected by
the onboardUser API with 400 INVALID_ARGUMENT. This matches the
opencode-antigravity-auth reference which only uses allowedTiers.
Co-Authored-By: Claude <noreply@anthropic.com>
- Store project IDs in composite refresh token format (refreshToken|projectId|managedProjectId)
- Add parseRefreshParts() and formatRefreshParts() for token handling
- Extract and persist subscription tier during project discovery
- Fetch subscription in blocking mode when missing from cached accounts
- Fix conditional duetProject setting to match reference implementation
- Export parseTierId() for reuse across modules
Co-Authored-By: Claude <noreply@anthropic.com>
Consolidate strategy configuration by moving STRATEGY_LABELS from
strategies/index.js to constants.js. Update startup banner to
dynamically display strategy options from SELECTION_STRATEGIES.
Co-Authored-By: Claude <noreply@anthropic.com>
Refactor account selection into a strategy pattern with three options:
- Sticky: cache-optimized, stays on same account until rate-limited
- Round-robin: load-balanced, rotates every request
- Hybrid (default): smart distribution using health scores, token buckets, and LRU
The hybrid strategy uses multiple signals for optimal account selection:
health tracking for reliability, client-side token buckets for rate limiting,
and LRU freshness to prefer rested accounts.
Includes WebUI settings for strategy selection and unit tests.
Co-Authored-By: Claude <noreply@anthropic.com>
Set functionCallingConfig.mode = 'VALIDATED' when using Claude models
to ensure strict parameter validation, matching opencode-antigravity-auth.
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: preserve whitespace-only chunks in SSE stream
Fixes issue #138 where Claude models would swallow spaces between words because whitespace-only chunks (e.g., " ") were being filtered out as empty.
Changes:
- Modified sse-streamer.js to only skip truly empty strings (""), preserving strings that contain only whitespace.
- Added regression test case in tests/test-streaming-whitespace.cjs to verify whitespace preservation.
* test: add streaming whitespace regression test to main suite
---------
Co-authored-by: walczak <walczak@ial.ruhr>
Some accounts have paidTier.id = "free-tier" but allowedTiers includes
"standard-tier", meaning they can use Pro-level quotas even though their
Google One subscription doesn't grant Antigravity benefits.
New priority:
1. paidTier - if Pro/Ultra, use immediately
2. allowedTiers - if paidTier is free, check for higher tiers
3. currentTier - fallback if still unknown
4. allowedTiers default - final fallback
This fixes accounts that show as "free" but actually have Pro access
through programs other than Google One AI.
Co-Authored-By: Claude <noreply@anthropic.com>
- Use raw API tier IDs (e.g., 'free-tier', 'g1-pro-tier') consistently
- Pass DEFAULT_PROJECT_ID for non-free tiers during onboarding
- Fix free tier detection to use .includes('free') instead of exact match
- Upgrade onboarding error logs from debug to warn for visibility
- Add debug logging for onboarding request parameters
Fixes onboarding failures for Pro/Ultra accounts that were previously
failing because the API requires a project ID for paid tier onboarding.
Co-Authored-By: Claude <noreply@anthropic.com>
- Use parseTierId() consistently for all tier sources (paidTier, currentTier, allowedTiers)
- Add tierSource tracking to identify which field tier was detected from
- Add verbose debug logging to show all tier-related fields from loadCodeAssist API
- Fix onboarding to prioritize paidTier > currentTier > allowedTiers (consistent with model-api.js)
This helps diagnose issues where Pro/Ultra accounts are incorrectly detected as free.
Fixes#128
Co-Authored-By: Claude <noreply@anthropic.com>
When all accounts are rate-limited for > 2 minutes, now attempts
model fallback (if enabled) before throwing the error. This allows
quota-exhausted accounts to gracefully fall back to alternate models.
Co-Authored-By: Claude <noreply@anthropic.com>
- Pass project ID to fetchAvailableModels for accurate per-project quota
- Treat missing remainingFraction with resetTime as 0% (exhausted)
- Fix double-escaped regex in rate-limit-parser.js (\\d -> \d)
- Use ANTIGRAVITY_HEADERS for loadCodeAssist consistency
- Store actual reset time from API instead of capping at default
- Add getRateLimitInfo() for detailed rate limit state
- Handle disabled accounts in rate limit checks
Fixes issue where free tier accounts showed 100% quota but were actually exhausted.
Co-Authored-By: Claude <noreply@anthropic.com>
When loadCodeAssist returns no project, automatically call onboardUser API
to provision a managed project. This handles first-time setup for new accounts.
Co-Authored-By: Claude <noreply@anthropic.com>
When Claude Code strips thinking signatures it doesn't recognize,
the proxy would drop unsigned thinking blocks, causing the error
"Expected thinking but found text". This fix detects unsigned
thinking blocks and triggers recovery to close the tool loop.
Co-Authored-By: Claude <noreply@anthropic.com>
The tier detection was incorrectly showing Pro accounts as "free" due to:
1. paidTier field being flaky/missing from some API responses
2. standard-tier not being recognized as a Pro tier
3. No fallback to allowedTiers when currentTier is missing
Changes:
- Add parseTierId() helper to centralize tier ID parsing
- Recognize "standard-tier" as Pro (Gemini Code Assist paid tier)
- Add fallback chain: paidTier > currentTier > allowedTiers
- Return "unknown" instead of incorrectly defaulting to "free"
Fixes#121
Co-Authored-By: Claude <noreply@anthropic.com>
When multiple tool_results contain images, the inlineData parts were
being interleaved between functionResponse parts, breaking Claude's API
requirement that functionResponse parts be consecutive.
Co-Authored-By: Claude <noreply@anthropic.com>
Replace gpt-tokenizer with model-specific official tokenizers:
- Claude models: @anthropic-ai/tokenizer (official Anthropic tokenizer)
- Gemini models: @lenml/tokenizer-gemini (GemmaTokenizer)
Changes:
- Add @anthropic-ai/tokenizer and @lenml/tokenizer-gemini dependencies
- Remove gpt-tokenizer dependency
- Update count-tokens.js with model-aware tokenization
- Use getModelFamily() to select appropriate tokenizer
- Lazy-load Gemini tokenizer (138MB) on first use
- Default to local estimation for all content types (no API calls)
Tested with all supported models:
- claude-sonnet-4-5, claude-opus-4-5-thinking, claude-sonnet-4-5-thinking
- gemini-3-flash, gemini-3-pro-low, gemini-3-pro-high
- Add ensureInitialized() call before count_tokens handler
- Use hybrid approach: local estimation for text, API for images/docs
- This prevents "No accounts available" error on first request
Switch from local estimation (gpt-tokenizer) to API-based counting
via Google Cloud Code API for accurate token counts. Falls back to
local estimation if API call fails.
Add Anthropic-compatible token counting endpoint using hybrid approach:
- Local estimation with gpt-tokenizer for text content (~95% accuracy)
- API-based counting for complex content (images, documents)
- Automatic fallback to local estimation on API errors
This resolves warnings in LiteLLM and other clients that rely on
pre-request token counting.
- Move TEST_MODELS and DEFAULT_PRESETS to src/constants.js as single source of truth
- Update test-models.cjs helper to use dynamic import from constants
- Make getTestModels() and getModels() async functions
- Update all test files to await async model config loading
- Remove duplicate THINKING_MODELS and getThinkingModels() from test helper
- Make thinking tests more lenient for Gemini (doesn't always produce thinking blocks)
Co-Authored-By: Claude <noreply@anthropic.com>
- Cooldown now caps API-provided reset times instead of being a fallback
- Fixed misleading UI descriptions for cooldown settings
- Removed unused cooldownDurationMs from settings object
- Updated default fallback values in frontend to 10s
Co-Authored-By: Claude <noreply@anthropic.com>
- Add separate LOAD_CODE_ASSIST_ENDPOINTS (prod first) and
LOAD_CODE_ASSIST_HEADERS (google-api-nodejs-client User-Agent)
- Add duetProject to metadata for project discovery
- Silent fallback when API returns success but no project
(matches opencode-antigravity-auth behavior)
- Only warn when all endpoints fail with actual errors
Fixes#114, addresses discussion #113
Co-Authored-By: Claude <noreply@anthropic.com>
