Skip to content
这个页面有中文版。
切换到中文
API Pricing

Anthropic: Claude Haiku 5.5 API pricing on OpenRouter

id
anthropic/claude-haiku-5.5
anthropic
First seen 2026-11-07 13:53 UTC
Open on openrouter.ai/anthropic/claude-haiku-5.5

RouteOfficial partnervia Anthropic + Claude Platform on AWS + Amazon Bedrock + 2 more

Basis: OpenRouter endpoints: Anthropic (anthropic): official; Claude Platform on AWS (claude-on-aws): official; Amazon Bedrock (amazon-bedrock): partner; Azure (azure): partner; Google (google-vertex): partner Archived copy

Current prices

Per million input tokens2

VariantUSDPlatform’s own wordingSince
via Google · global$0.1/M inprompt: 0.0000001 / token (endpoint google-vertex/global)
Prompt > 100Kvia Google · global$0.5/M inprompt: 0.0000005 / token (>100000; endpoint google-vertex/global)

Per million output tokens2

VariantUSDPlatform’s own wordingSince
via Google · global$0.5/M outcompletion: 0.0000005 / token (endpoint google-vertex/global)
Prompt > 100Kvia Google · global$2.5/M outcompletion: 0.0000025 / token (>100000; endpoint google-vertex/global)

Per million cached input tokens (read)2

VariantUSDPlatform’s own wordingSince
via Google · global$0.01/M cache readinput_cache_read: 0.00000001 / token (endpoint google-vertex/global)
Prompt > 100Kvia Google · global$0.05/M cache readinput_cache_read: 0.00000005 / token (>100000; endpoint google-vertex/global)

Per million cache-write tokens4

VariantUSDPlatform’s own wordingSince
5-min cachevia Google · global$0.125/M cache writeinput_cache_write: 0.000000125 / token (endpoint google-vertex/global)
1-h cachevia Google · global$0.2/M cache writeinput_cache_write_1h: 0.0000002 / token (endpoint google-vertex/global)
5-min cachePrompt > 100Kvia Google · global$0.625/M cache writeinput_cache_write: 0.000000625 / token (>100000; endpoint google-vertex/global)
1-h cachePrompt > 100Kvia Google · global$1/M cache writeinput_cache_write_1h: 0.000001 / token (>100000; endpoint google-vertex/global)

Per run (default parameters)1

VariantUSDPlatform’s own wordingSince
Web searchvia Google · global$0.01/runweb_search: 0.01 (endpoint google-vertex/global)

Change log

No changes yet. A platform’s first run sets the baseline; changes appear from the next one.

Earlier versions

Model · what it coversUSDPlatform’s own wordingSince
$0.1/M inprompt: 0.0000001 / token2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
Prompt > 100K$0.5/M inprompt: 0.0000005 / token (>100000)2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
$0.5/M outcompletion: 0.0000005 / token2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
Prompt > 100K$2.5/M outcompletion: 0.0000025 / token (>100000)2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
Cache read$0.01/M cache readinput_cache_read: 0.00000001 / token2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
Cache readPrompt > 100K$0.05/M cache readinput_cache_read: 0.00000005 / token (>100000)2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
Cache write (5 min)$0.125/M cache writeinput_cache_write: 0.000000125 / token2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
Cache write (1 h)$0.2/M cache writeinput_cache_write_1h: 0.0000002 / token2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
Cache write (5 min)Prompt > 100K$0.625/M cache writeinput_cache_write: 0.000000625 / token (>100000)2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
Cache write (1 h)Prompt > 100K$1/M cache writeinput_cache_write_1h: 0.000001 / token (>100000)2026-11-07 13:53 UTC → 2026-11-11 16:55 UTC
Web search$0.01/runweb_search: 0.012026-11-07 13:53 UTC → 2026-11-11 16:55 UTC

Where these prices come from

Read from the platform’s public pages and APIs. Every read is archived, so any price here can be checked against what the platform showed at the time.