Claude Haiku 5.5
Anthropic's 7 Oct 2026 small model, built for classification, extraction, routing and subagent work, and the first Haiku with an adjustable effort setting. $0.10/$0.50 per 1M tokens for prompts up to 100K tokens, rising to $0.50/$2.50 for prompts over 100K (cache reads $0.01 / $0.05) — Anthropic says that is 90% cheaper than Haiku 4.5 at the short tier, and that Sonnet 5.5 cache reads fell from $0.20 to $0.10 the same day. The price shown is the short-prompt tier; the long-prompt tier applies to the whole request. 1M context, 128K maximum output, adaptive thinking. Artificial Analysis Intelligence Index v5.0.1 scores it 36 at max effort, #1 of 34 comparable models, at $0.41 per Index task and 255 tokens/s — but with 298M output tokens against a 97M median and 185.68s to the first answer token, so it is a throughput model, not an interactive one. Anthropic's own table has it ahead of GPT-6 Luna on GDPval-AA (1620 vs 1437), OSWorld 2.1 (72.4% vs 48.9%) and Terminal-Bench 4.0 (39.2% vs 16.4%); it still trails Sonnet 5.5 and Opus 5.5 on complex agentic coding.
1M
tokens
128K
tokens
$0.1
per 1M tokens
$0.5
per 1M tokens
255
tokens/sec
Oct 2026
2026-10-07
$0.30
per 1M tokens
303.3
quality per $
Capabilities
Benchmarks
Price vs Quality
Context Window in Context
- ▸ Claude Haiku 5.51M
- GPT-4.11M
- Gemini 2.5 Pro1M
- Gemini 2.0 Flash1M
- Llama 4 Maverick1M
- GPT-5.51M
- GPT-5.5 Pro1M
- Claude Opus 4.71M
Not on our gateway yet
Swfte Connect does not hold a key for Claude Haiku 5.5, so there is nothing to run here. Add your own provider key and call it with the same API as every other model in this directory.
curl https://api.swfte.com/agents/v2/gateway/chat/completions \
-H "Authorization: Bearer $SWFTE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "anthropic:claude-haiku-5-5",
"messages": [{"role": "user", "content": "Hello"}]
}'About Claude Haiku 5.5
Claude Haiku 5.5 is a fast AI model by Anthropic, released on October 7, 2026. It supports a context window of 1000K tokens and can generate up to 128K output tokens.
At $0.1 per million input tokens and $0.5 per million output tokens, its blended cost of $0.30/1M tokens makes it one of the most affordable models available. Its value score of 303.3 reflects the balance of quality and cost.
Using Claude Haiku 5.5 with Swfte
Access Claude Haiku 5.5 through Swfte Connect, our unified LLM gateway. Connect gives you a single API for 50+ models, with automatic routing, cost optimization, and fallback handling. You can also try Claude Haiku 5.5 in our AI Playground before integrating.