Claude Haiku 5.5
Hosted by a provider reached through SayGM. The route is sealed in transit and in use; the provider receives the prompt under its API terms.
Released
Oct 2026Knowledge cutoff
Jun 2026Best reported price on SayGM
-10.7%In
Cached
Out
per Mtok*
Call Claude Haiku 5.5 from the API
Create an API key and set it as SAYGM_API_KEY in your terminal.
curl "https://api.saygm.com/v1/messages" \
-H "x-api-key: $SAYGM_API_KEY" \
-H "Content-Type: application/json" \
-H "anthropic-version: 2023-06-01" \
-d '{
"model": "claude-haiku-5-5",
"max_tokens": 256,
"messages": [
{
"role": "user",
"content": "Hello!"
}
]
}'SayGM vs. Anthropic list price
SayGM's best reported rate for claude-haiku-5-5 next to Anthropic's own published list price — 10.7% off.
| Dimension | SayGM price | Anthropic list price |
|---|---|---|
| Input | $0.0893per Mtok | $0.10 |
| Output | $0.4465per Mtok | $0.50 |
| Cache read | $0.00893per Mtok | $0.01 |
| Cache write, 5 minute | $0.111625per Mtok | $0.125 |
| Cache write, 1 hour | $0.1786per Mtok | $0.20 |
| Long-context inputabove 100,000 input tokens | $0.4465per Mtok | $0.50 |
| Long-context outputabove 100,000 input tokens | $2.2325per Mtok | $2.50 |
| Long-context cache readabove 100,000 input tokens | $0.04465per Mtok | $0.05 |
| Long-context cache write, 5 minuteabove 100,000 input tokens | $0.558125per Mtok | $0.625 |
| Long-context cache write, 1 hourabove 100,000 input tokens | $0.893per Mtok | $1.00 |
Recent effective price
What buyers have actually paid on this model recently, across every provider and cache hits.
Input
$0.09
$0.10Cached input
$0.01
Output
$0.45
$0.50How the effective price is calculated
Requests are load balanced across every provider serving this model, each offering its own rate, so what you pay is a blend rather than a single number. These are the blended rates actually paid — priced at what a provider offered, capped at list.
Measured over 5 windows: 553 requests and 2,170,237 input-side tokens. Most recent window closed 2026-10-10 03:15 UTC.
Who's serving this model
Every provider currently serving this model, ranked by discount.
| Provider | Discount | In | Out |
|---|---|---|---|
| Provider 1 | 10.7% off | $0.09 | $0.45 |
| Provider 2 | 10.7% off | $0.09 | $0.45 |
| Provider 3 | 10.6% off | $0.09 | $0.45 |
| Provider 4 | 10.6% off | $0.09 | $0.45 |
| Provider 5 | 10.5% off | $0.09 | $0.45 |
| Provider 6 | 10.5% off | $0.09 | $0.45 |
| Provider 7 | 10.4% off | $0.09 | $0.45 |
| Provider 8 | 10.4% off | $0.09 | $0.45 |
| Provider 9 | 10.3% off | $0.09 | $0.45 |
| Provider 10 | 10.3% off | $0.09 | $0.45 |
| Provider 11 | 10.3% off | $0.09 | $0.45 |
| Provider 12 | 10.2% off | $0.09 | $0.45 |
| Provider 13 | 10.2% off | $0.09 | $0.45 |
| Provider 14 | 10.1% off | $0.09 | $0.45 |
| Provider 15 | 10.1% off | $0.09 | $0.45 |
Performance
Measured on the requests SayGM routed to this model.
- Throughput
- 181.9 tok/sover 2,900 requests
- Time to first token
- 0.89 sover 2,883 requests
- Request success rate
- 100.0% over 102 requests
Benchmarks
Fixed-seed suites SayGM runs against this model.
No benchmark run has been published for this model.
More from Anthropic
Explore the Anthropic lab and compare related models.
Frequently asked questions
How much does claude-haiku-5-5 cost on SayGM?
The best reported rate for claude-haiku-5-5 on SayGM is $0.09 per million input tokens and $0.45 per million output tokens. A request above 100,000 input tokens bills at SayGM's long-context rate instead.
How much cheaper is SayGM than Anthropic list price?
SayGM's best reported rate for claude-haiku-5-5 is 10.7% off Anthropic's list price.
Which SDK do I use for claude-haiku-5-5 on SayGM?
The Anthropic SDK. Set its base URL to https://api.saygm.com and your existing Messages API code (/v1/messages) calls claude-haiku-5-5 as-is.
Related reading
- SayGM Desktop: One AI Assistant App for Every ModelClaude, GPT, Gemini and more in one desktop app, always cheaper than list price. Meet SayGM Desktop: an AI assistant that works on your files, runs tasks on a schedule and asks before it acts.
- OpenAI API Pricing 2026: GPT-6.1 Sol, Astra and LunaGPT-6.1 Sol costs $2.00 per million input tokens on OpenAI. Here's what it and every other GPT model costs direct, through Azure, Bedrock, OpenRouter and Requesty, and on SayGM, where every GPT model is 22% below OpenAI's standard rate.
- Get Paid to Refer Developers to SayGM's AI GatewaySayGM's new affiliate program pays you for every developer you bring to its AI gateway: up to $1 for every $1 they spend. Rewards come from a shared pool of Subnet 28 emissions, split by each affiliate's share of referred revenue, with every round and payout recorded on chain. It's a competition, not a commission, built on proof of customers.