Claude Haiku 5.5
ModelAPIProprietary model (hosted access only)
What it is
Anthropic's small, fast Claude model. It is positioned for high-volume work like classification and routing, which suits first-line support triage.
Best for
#1 of 3Model for Build a customer-support agent
Total 7.4 / 10
- Task quality (15%)7 / 1075.3% τ²-Airline, within ~3pp of Sonnet 5.5.
- Cost per typical job run (25%)9 / 10$0.012/task on τ²-Airline; $0.10/$0.50 per MTok.
- Latency / speed (15%)6 / 101.6 min/task on τ²-Airline (multi-turn task, not per-reply latency).
Evidence:0014 (source: OpenRouter)
- Tool-calling & structured-output reliability (25%)8 / 10
Evidence:0014 (source: OpenRouter)
- Context & modality support (10%)8 / 101M context.
Evidence:0017 (source: Anthropic)
- Openness / deployability (10%)4 / 10Closed weights; API + Bedrock + Google Cloud.
Evidence:0016 (source: Anthropic)
Strengths
- Low price: $0.10 input / $0.50 output per million tokens (prompts up to 100K).0016 (source: Anthropic)
- 1M-token context window and 128K max output.0017 (source: Anthropic)
- Positioned for high-volume classification and routing.0017 (source: Anthropic)
Limits
- Proprietary, API-only model; cannot be self-hosted.0016 (source: Anthropic)
- Public τ²-bench figures for it are listed as estimated/low confidence, not measured.0015 (source: benchgap)
Pricing
Paid
$0.10 input / $0.50 output per MTok (prompts ≤100K).0016 (source: Anthropic)
As of . Check the vendor's page before buying.
License
Proprietary model (hosted access only).0160 (source: Anthropic)0174 (source: Anthropic)0159 (source: Anthropic)
Checked .
Alternatives in the same layer
Disclosures
- None