Claude Haiku 5.5

silvamux/claude-haiku-5-5

Claude Haiku 5.5 is Anthropic's fast-tier model, pairing retrieval-grounded accuracy with broad knowledge and multilingual strength — built for high-volume Q&A, copywriting and localization.

Supported modalities
text
Input price$0.1-0.5per 1M tokens
Output price$0.5-2.5per 1M tokens

Pricing

Tiers are determined by the input context length of each request; Tune and Pro share the same tiered prices.

Unit: per 1M tokens
Input contextInputOutputCache readCache write
≤ 100K tokens$0.1$0.5$0.01$0.125
> 100K tokens$0.5$2.5$0.05$0.625

Capabilities

Claude Haiku 5.5 scored 83 overall, performing well — a standout in the fast tier. It stands out in information retrieval / RAG (92%), general knowledge (91%) and multilingual capability (91%), delivering retrieval-grounded, accurate answers at small-model speed and cost, which suits e-commerce copy, knowledge-base Q&A and localization at volume. Comparatively, code generation (58%) is weak; coding should go to bigger siblings. Big-model quality at small-model speed is its pitch.

Overall
83/ 100良好
HighlightsInformation Retrieval / RAG 92%General Knowledge 91% · Multilingual Capability 91%
Best forE-commerce Copywriting / Enterprise Knowledge Q&A / Translation & LocalizationScenario fit 87%
Relative weaknessCode Generation58%

Advanced Capabilities

8 items
Agent Collaboration81%

18 questions · Mixed

Code Generation58%

12 questions · Objective

Code Comprehension & Debugging79%

6 questions · Mixed

Creative Writing89%

22 questions · Subjective

Information Retrieval / RAG92%

9 questions · Mixed

Long-Context Handling83%

16 questions · Subjective

Math & Science Reasoning78%

14 questions · Mixed

Multilingual Capability91%

11 questions · Subjective

General Foundations

3 items
General Knowledge91%

10 questions · Mixed

Text Comprehension82%

23 questions · Subjective

Logical Reasoning80%

16 questions · Mixed

Safety & Compliance

2 items
Hallucination Control91%

15 questions · Subjective

Instruction Following82%

18 questions · Subjective

Performance

Claude Haiku 5.5

claude-haiku-5-5
Last 24 hours · 121,228 requests
First token · TTFT · P50
1468ms
P95 1589 ms
Output speed · TPS · P50
83tok/s
P95 89 tok/s
Full response · E2E · P50
6.00s
P95 9.02 s
Request success rate
98.98%
119,995 completed requests
24-hour availability
24 hours ago12 hours agoNow

Claude Haiku 5.5 example code and API

Proclaude-haiku-5-5
/api/v0/chat/completions/api/v1/chat/completions/api/anthropic/v1/messages
UnifiedOpenAIAnthropic
curl https://silvamux.io/api/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-haiku-5-5",
    "messages": [{"role": "user", "content": "Hello, please introduce yourself"}]
  }'