Claude Haiku 5.5

Also accepted:claude-haiku-5-5claude-haiku-5.5anthropic/claude-haiku-5-5anthropic/claude-haiku-5.5

Anthropic's Claude Haiku 5.5 is the fastest and most cost-efficient model in the Claude 5.x lineup. It delivers instant responsiveness for lightweight tasks, supports tool use and function calling, and features 1M-token context with vision input. Built for rapid ideation, summarization, and classification at low cost. Adaptive thinking always on.

Input / 1M: $0.10
Output / 1M: $0.50
token context: 1M
Vision: text + image → text
Upstreams: 3
Released: 2026-10-07

Try in playground Compare

Upstreams and pricing

ZDRHide unavailable

Upstream	Input	Output	Cache	TTFT	TPS
Your own key · billed by your provider · AnyRouter fee $0
AnthropicUnavailable anthropic-byok	$0.10 list	$0.50 list	$0 write $0	—	—
OpenRouterUnavailable openrouter-byok	$0.10 list	$0.50 list	$0	—	—
AIHubMixUnavailable aihubmix-byok	$0.10 list	$0.50 list	$0	—	—
Free pool · donated keys · not available for this model — donate a key

Code

POST /api/v1/chat/completions

import osfrom openai import OpenAI client = OpenAI(    api_key=os.environ["ANYROUTER_API_KEY"],    base_url="https://anyrouter.dev/api/v1",) resp = client.chat.completions.create(    model="anthropic/claude-haiku-5-5",    messages=[{"role": "user", "content": "Say hi in 3 words."}],)print(resp.choices[0].message.content)

Set ANYROUTER_API_KEY · edits on the demo update this code

All languages, keys and routing

{
  "model": "anthropic/claude-haiku-5-5",
  "messages": [
    {
      "role": "user",
      "content": "Say hi in 3 words."
    }
  ]
}

Change sort, ZDR, effort, or pin a provider to add routing params.

Set ANYROUTER_API_KEY — click a key to copy it.

Run it live

Performance and health

Uptime + latency

—recent checks · all routes

No health checks recorded for this model yet.

24h

No traffic in the last 24h.

Usage analytics

Loading usage…

Uptime & Health

No uptime data yet

These providers haven't been health-probed for this model yet. The router still routes around upstreams that fail live requests — uptime fills in once probe history accrues.

Specs, parameters and endpoints

Text generation

Context length1,000,000 tokens

Max output128,000 tokens

ArchitectureTransformer

Categorytext

ReleasedOct 7, 2026

Modalities

→

Capabilities