Migration assistant

Replacing Claude Haiku 3.5 (<a class="inline-link" href="/docs/en/about-claude/model-deprecations">retired, except on Bedrock and Google Cloud</a>)?

Already retired — replacement required. Ranked replacement candidates, each shown with exactly what changes if you switch. Same-provider matches surface first because switching cost is lower; capability regressions are flagged in red.

AN
Migrating from

Claude Haiku 3.5 (<a class="inline-link" href="/docs/en/about-claude/model-deprecations">retired, except on Bedrock and Google Cloud</a>)

Anthropic · Retired
Input: $0.8/M Output: $1.6/M Context:
View detail →
#1 RECOMMENDED

Claude Haiku 3.5

Anthropic
same provider
View full detail →
What changes if you switch
Field Claude Haiku 3.5 (<a class="inline-link" href="/docs/en/about-claude/model-deprecations">retired, except on Bedrock and Google Cloud</a>) Claude Haiku 3.5 Impact
Input price $0.8/M $0.8/M Same price
Output price $1.6/M $4/M +150% more expensive
Prompt caching ✓ supported ✓ supported Preserved
Lifecycle Retired Active Generally available
Other candidates
#2

Claude Haiku 4.5

Anthropic
same provider; 25% more expensive input
Details →
Input price: +25% more expensiveOutput price: +212% more expensiveLifecycle: Generally available
#3

Claude Fable 5

Anthropic
same provider; 1150% more expensive input
Details →
Input price: +1150% more expensiveOutput price: +525% more expensiveLifecycle: Generally available
#4

Claude Mythos 5 (limited availability)

Anthropic
same provider; 1150% more expensive input
Details →
Input price: +1150% more expensiveOutput price: +3025% more expensiveLifecycle: Generally available
#5

Claude Opus 4

Anthropic
same provider; 1775% more expensive input
Details →
Input price: +1775% more expensiveOutput price: +4588% more expensiveLifecycle: Generally available
Methodology

How candidates are ranked

Candidates are ranked by how much of the source model's profile each one preserves — weighing switching cost (same provider or model family surface first), capability parity (what you keep versus what you'd lose), context window, and pricing direction. A candidate that is itself deprecated is pushed down the list; one that's already retired is never recommended.

Ranking is purely algorithmic — no editorial weighting, no paid placement. Every value is pulled from each provider's own documentation; click any model name to see the source-linked detail.