[Bug] Unexpectedly High API Usage and Rate Limit Hits
Status Closed — not planned
Reported on v2.1.87
Maintainer reply None cached
Workaround ✓ Mentioned in thread ↓
Activity 11 comments · opened Mar 31, 2026 · closed May 24, 2026
Bug Description
The usage this week was absurdly high, I've hit limits much faster than before, I've read I'm not the only one and that it's caused by bugs
Environment Info
- Platform: darwin
- Terminal: tmux
- Version: 2.1.87
- Feedback ID: 42960fb3-fb3b-4b74-bc76-c8656ba90099
Errors
[{"error":"Error: NON-FATAL: Lock acquisition failed for /Users/cminds/.local/share/claude/versions/2.1.87 (expected in multi-process scenarios)\n at _p_ (/$bunfs/root/src/entrypoints/cli.js:2851:2138)\n at Vn6 (/$bunfs/root/src/entrypoints/cli.js:2851:1218)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-03-29T17:53:40.525Z"},{"error":"Error: Agent type 'brand-kit-gen' not found. Available agents: general-purpose, statusline-setup, Explore, Plan, claude-code-guide, code-simplifier:code-simplifier, huggingface-skills:AGENTS, feature-dev:code-reviewer, feature-dev:code-explorer, feature-dev:code-architect, plugin-dev:skill-reviewer, plugin-dev:agent-creator, plugin-dev:plugin-validator, py-docs, wp-testing, py-planning, wp-planning, wp-docs, py-testing, py-research, wp-research, datastar-expert, ux-reviewer, robin, alfred, openrouter-delegate, batman, gemini-analyzer\n at call (/$bunfs/root/src/entrypoints/cli.js:3184:4869)\n at ji1 (/$bunfs/root/src/entrypoints/cli.js:3922:7155)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-03-30T20:21:45.464Z"},{"error":"Error: 429 {\"type\":\"error\",\"error\":{\"type\":\"rate_limit_error\",\"message\":\"This request would exceed your account's rate limit. Please try again later.\"},\"request_id\":\"req_011CZappfYCdo7NdC7jY2Mva\"}\n at generate (/$bunfs/root/src/entrypoints/cli.js:11:51105)\n at makeRequest (/$bunfs/root/src/entrypoints/cli.js:50:5435)\n at processTicksAndRejections (native:7:39)","timestamp":"2026-03-31T08:15:23.412Z"},{"error":"Error: ensureToolResultPairing: repaired missing tool_result blocks (255 -> 255 messages). Message structure: [0] user; [1] assistant(id=msg_011HVGzHxA6JPiSW2UJz5PTy, tool_uses=[]); [2] user; [3] assistant(id=msg_01YA7LvKmwUwRCVyEdY3W6P3, tool_uses=[toolu_01Xy2MQvEyjUDwKaW9pBPoZW]); [4] user(tool_results=[toolu_01Xy2MQvEyjUDwKaW9pBPoZW]); [5] assistant(id=msg_01EG2U3dorSDiBLG3Pv8mYUo, tool_uses=[toolu_01EFQLJiE7rybwbScv9ukQQr,toolu_01HcyuqiRML4GKfxESv63Dpm,toolu_01GWWaazDVcWQJNYx4NkqqDv]); [6] user(tool_results=[toolu_01EFQLJiE7rybwbScv9ukQQr,toolu_01HcyuqiRML4GKfxESv63Dpm,toolu_01GWWaazDVcWQJNYx4NkqqDv]); [7] assistant(id=msg_01UzyeqhBgjY7VxcgwuFK4eZ, tool_uses=[toolu_01Qu1ML2zeALpLd6M6w91yb1,toolu_014eKaovcvurLXArAAP4ojZb,toolu_01FjM6fcSb3xA3okEna1VANj]); [8] user(tool_results=[toolu_01Qu1ML2zeALpLd6M6w91yb1,toolu_014eKaovcvurLXArAAP4ojZb,toolu_01FjM6fcSb3xA3okEna1VANj]); [9] assistant(id=msg_01LWUXegLuSV9Rjy8uDx4Si7, tool_uses=[]); [10] user; [11] assistant(id=msg_017kwSBzcM1vNGsyg3DQR7ar, tool_uses=[toolu_01MZXt1eBvDbpx4FUmwTgkNK,toolu_01YMwFXh4PZYdFEgJRAZqu1J]); [12] user(tool_results=[toolu_01MZXt1eBvDbpx4FUmwTgkNK,toolu_01YMwFXh4PZYdFEgJRAZqu1J]); [13] assistant(id=msg_018D9vwRUuHWKWEbGN13TPjt, tool_uses=[toolu_01R72iCWpw3weyoMMtXRjqvq]); [14] user(tool_results=[toolu_01R72iCWpw3weyoMMtXRjqvq]); [15] assistant(id=msg_01K3tceqQ2Z6mjvYTCGVH2Yw, tool_uses=[toolu_01XqmYy6F3QGQvZDi8WpmUGT,toolu_01CUJRrmFeit1iHjtVKbKQu1,toolu_013VQ8J3Sdf4NpAKQswGNhto]); [16] user(tool_results=[toolu_01XqmYy6F3QGQvZDi8WpmUGT,toolu_01CUJRrmFeit1iHjtVKbKQu1,toolu_013VQ8J3Sdf4NpAKQswGNhto]); [17] assistant(id=msg_01MkgYyZiiXxAZnovRh8G1TS, tool_uses=[toolu_01KS161S9EYgLheU1Mx1wcSC,toolu_01EAXeLmE25owmLzxTqURFqR,toolu_01GTkP4QLzy4HYharMqCJwon]); [18] user(tool_results=[toolu_01KS161S9EYgLheU1Mx1wcSC,toolu_01EAXeLmE25owmLzxTqURFqR,toolu_01GTkP4QLzy4HYharMqCJwon]); [19] assistant(id=msg_01XiEZM2wz1QYHKrB2Mniv77, tool_uses=[toolu_015XnNi5h7C69SHZiaTuF8Ac,toolu_01Gyhb33h3ttkLQNGbnjU34w,toolu_01YDivJQKqrU71hFxuMtxQPA]); [20] user(tool_results=[toolu_015XnNi5h7C69SHZiaTuF8Ac,toolu_01Gyhb33h3ttkLQNGbnjU34w,toolu_01YDivJQKqrU71hFxuMtxQPA]); [21] assistant(id=msg_01VC2Chtmtorb5BcMEMUyXgj, tool_uses=[toolu_01L7XDyXhX41UYTEcx6jjStS]); [22] user(tool_results=[toolu_01L7XDyXhX41UYTEcx6jjStS]); [23] assistant(id=msg_01W9Ta2imiRf8JjEGAfL8Eam, tool_uses=[toolu_01PGxnMGqqq4Fge6chwXMev9,toolu_012LYSfuYP36vbpQhxExsrVr,toolu_01RnEnAzKA5U8pH1cbhGWeHL,toolu_01QWGp1Vn9pa9tYMZnyuPoK4,toolu_01LeWfr1Lv2ZPVGPexXei4je,toolu_01SMpiHZY4yJx5FvFni3fXwA,toolu_0195HztKrMWL3o61La5E78uz]); [24] user(tool_results=[toolu_01PGxnMGqqq4Fge6chwXMev9,toolu_012LYSfuYP36vbpQhxExsrVr,toolu_01RnEnAzKA5U8pH1cbhGWeHL,toolu_01QWGp1Vn9pa9tYMZnyuPoK4,toolu_01LeWfr1Lv2ZPVGPexXei4je,toolu_01SMpiHZY4yJx5FvFni3fXwA,toolu_0195HztKrMWL3o61La5E78uz]); [25] assistant(id=msg_01D6SmGoxxiL3me4VR97UQQY, tool_uses=[toolu_01MYfeKRp6yo2j94tkjLR9yF]); [26] user(tool_results=[toolu_01MYfeKRp6yo2j94tkjLR9yF]); [27] assistant(id=msg_01Pysz6Jy6hawiKTGMsqZ9jF, tool_uses=[toolu_01BrydDqQEPVzjvaWHTrtSA2]); [28] user(tool_results=[toolu_01BrydDqQEPVzjvaWHTrtSA2]); [29] assistant(id=msg_01PN241zPvRwx5P4v8Mnx31q, tool_uses=[toolu_01UDqJ4fyddyLLCm1NchodW5]); [30] user(tool_results=[toolu_01UDqJ4fyddyLLCm1NchodW5]); [31] assistant(id=msg_01N9NPodTEFYetrjVHDBjpmG, tool_uses=[toolu_01JUPBRwNEMPLngr98Q5F6Jm,toolu_01Dp4AKRR9Aaa4tPcZx83bGx]); [32] user(tool_results=[toolu_01JUPBRwNEMPLngr98Q5F6Jm,toolu_01Dp4AKRR9Aaa4tPcZx83bGx]); [33] assistant(id=msg_019W8dtVU8trC75j834g1zzV, tool_uses=[toolu_01PjfYWoHYgtNYjSJuHS2pY7]); [34] user…
Note: Content was truncated.
11 Comments
Found 3 possible duplicate issues:
This issue will be automatically closed as a duplicate in 3 days.
🤖 Generated with Claude Code
Same. Max 20 ($200/mo), v2.1.89. April 1: 100% in ~70 min after reset. Light usage only.
Full report: #41788
Related: #38335, #38239, #41663, #41506
I've been experiencing the same issue on Max 20 ($200/mo) — rate limit 100% exhausted in ~70 minutes.
After setting up a monitoring proxy using the official
ANTHROPIC_BASE_URLenv var, I identified two cache bugs as the root cause (#40524, #34629) and measured the impact: cache read ratio dropped to 4.3%, meaning ~20x token inflation per turn. After applying workarounds it stabilized at 89-99%.Full analysis with per-request measured data, safe workarounds, and community references (including cc-cache-fix): https://github.com/ArkNill/claude-code-cache-analysis
For me, 01/04/2026, this morning: rate limit 100% exhausted in ~50 minutes. I waited 4 hours and the rate limit was 100% exhausted again with a single prompt.
Follow-up — precautions (April 2, 2026)
For anyone landing here, additional behaviors that compound the cache bugs:
Avoid:
--resume(replays full history as billable input),/dream&/insights(background drain), v2.1.89 (cache bug + terminal regression)Careful with: sub-agents/Haiku (0% cache read), multiple terminals, large CLAUDE.md
Recommended: v2.1.81 fixed, fresh sessions, proxy monitoring
Ref: https://github.com/ArkNill/claude-code-cache-analysis
Update (April 2): v2.1.90 has significantly improved cache efficiency — benchmark shows 95-99% cache read in stable sessions (both npm and standalone installations).
If you're still affected:
claude update(ornpm install -g @anthropic-ai/claude-code)"DISABLE_AUTOUPDATER": "1"to~/.claude/settings.jsonenv section--resume(still broken)Note: server-side quota issues (org-level pool sharing, accounting mismatches) remain unresolved — the above fixes the client-side cache drain only.
Benchmark data: https://github.com/ArkNill/claude-code-cache-analysis
April 3 update: v2.1.91 fixes the cache regression that caused the worst drain. If you are still hitting limits after updating, there are additional unfixed mechanisms: a 200K tool result budget cap, a client-side false rate limiter, and silent context stripping — all confirmed via proxy testing. Anthropic acknowledged peak-hour tightening on X (Lydia Hallie) but stated "none were over-charging you." Measured data and analysis: claude-code-cache-analysis
My Claude says issue #41506 has the most comprehensive analysis from two users who measured the problem from both client-side and server-side, including a timeline of which versions are affected and specific workarounds.
Context bloat compounds silently — each turn re-sends accumulated tool outputs, progress entries, and metadata. Cozempic strips this automatically so you get more turns per rate limit window. Lossless — nothing meaningful is removed.
pip install cozempic && cozempic initFeedback welcome.
Closing for now — inactive for too long. Please open a new issue if this is still relevant.
This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.