Claude API Priority Tier SLAAnthropic · AI models · Anthropic · AI models
AI modelsChecked 15h ago
Promise vs reality
Downtime used against the allowance
4.4 d of major incidents, 2.4x the allowance44 h allowed in 365 days
18 major and 264 minor incidents. 22 d degraded (not counted).
Major incident minutes per month
Allowed per month: 3.6 h
- Promised
- 99.5%
- Observed, 365d
- 98.8%
- Major incident time
- 4.4 d
- Credit on first breach
- n/a
Commitments and credits
- Priority Tier (existing commitments only)Organizations with an existing Priority Tier capacity commitment (input TPM + output TPM, 1/3/6/12 months, specific model). No longer sold.99.5%
Monthly uptime below, credit of the bill
No credit schedule published for this tier.
Terms
- Measured
- Stated as a target, not a contractual SLA; measurement window not published.
Not covered
- Target only; no service-credit schedule published
- Requests beyond committed capacity fall back to standard tier
- Not supported on Claude Fable 5.1, Mythos 5.1/5/Preview, Opus 5.5, Opus 5, Sonnet 5
- Requests exceeding regular rate limits are declined
Priority Tier targets 99.5% uptime with prioritized computational resources. Requests beyond your committed capacity automatically fall back to standard tier.
Page banner (fetched 2026-09-23): 'Priority Tier capacity commitments are no longer available for purchase. Organizations with an existing commitment can continue to use Priority Tier through their contract end date'. For guaranteed capacity it points to contact sales. The 99.5% figure is worded as a target, not an SLA with credits.
Evidence: incidents, last 365 days
- Sep 22, 2026Elevated errors for multiple modelsclaude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Sep 15, 2026Intermittent error spikes for Claude Mythos 5.1 and Claude Fable 5.1claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Sep 11, 2026Elevated errors for Claude Mythos 5.1 and Claude Fable 5.1claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Sep 10, 2026Elevated latency on the Claude API for some usersClaude API (api.anthropic.com), Claude CoworkMinor
- Sep 3, 2026Elevated errors for multiple modelsclaude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Sep 3, 2026Elevated errors for Claude Sonnet 5claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Sep 2, 2026Elevated errors for Claude Sonnet 5claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Aug 24, 2026Elevated errors for multiple modelsclaude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Aug 20, 2026Elevated errors on requests to multiple modelsclaude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Aug 20, 2026Elevated errors on Google connectorsclaude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Aug 19, 2026Degraded performance for Claude Opus 5 and Claude Haiku 4.5claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Aug 18, 2026Degraded performance for multiple modelsclaude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Aug 17, 2026Degraded performance for Claude Opus 5, Claude Sonnet 5claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Aug 16, 2026Service disruption on Claude servicesclaude.ai, Claude Console (platform.claude.com), Claude API (api.anthropic.com), Claude Code, Claude CoworkOutage
- Aug 14, 2026Service disruption on Claude servicesClaude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Aug 13, 2026Elevated errors for Claude Mythos 5, Claude Fable 5, and Claude Sonnet 5claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkDegraded
- Aug 12, 2026Degraded performance for multiple modelsclaude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Aug 5, 2026Degraded performance for Claude Opus 5claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Aug 5, 2026Degraded performance of multiple modelsclaude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Aug 4, 2026Elevated errors on Claude Sonnet 5claude.ai, Claude Console (platform.claude.com), Claude API (api.anthropic.com), Claude CodeMinor
- Aug 4, 2026Elevated errors on Claude Sonnet 5claude.ai, Claude Console (platform.claude.com), Claude API (api.anthropic.com), Claude CodeMinor
- Aug 3, 2026Degraded performance on Claude Sonnet 5claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Aug 3, 2026Error rates across multiple modelsclaude.ai, Claude Console (platform.claude.com), Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Jul 31, 2026Degraded performance on Claude Sonnet 5claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
- Jul 30, 2026Degraded performance on Claude Opus 4.8claude.ai, Claude API (api.anthropic.com), Claude Code, Claude CoworkMinor
SLA changes
- Sep23Claude API Priority TierPriority Tier moved from a purchasable offer ('Get started with Priority Tier ... Higher availability: Target 99.5% uptime', still present in the 2025-11-11 snapshot) to 'no longer available for purchase' for new commitments. The 99.5% target is unchanged. Exact change date unknown; after 2025-11-11, observed 2026-09-23.Source
Other ai models SLAs
ServiceDowntime vs allowed
Amazon BedrockAWS99.9% / 99.178%
Vertex AI (Gemini API on Vertex)Google Cloud99.9% / 100%
Azure OpenAI in Foundry ModelsMicrosoft Azure99.9% / n/a
Claude API Priority TierAnthropic99.5% / 98.8%
Cohere API (SaaS: Command, Embed, Rerank)Coherenone / 100%
Hugging Face Inference EndpointsHugging Facenone / n/aNo public SLA
Promised in the SLAObserved on the vendor's status pagePartial historyShortfall
Summaries of published SLAs; the contract you sign governs. Logos via logo.dev; trademarks belong to their owners.