Anthropic
Every Anthropic model on Synthorai: one endpoint, live prices.
11 models · chat / code / tools / thinking · from $1/M input
Why Anthropic via Synthorai
- Bring your own key. Bring your own Anthropic key, or your AWS Bedrock / GCP Vertex credentials. Zero markup, free.
- Zero prompt retention. Zero prompt retention at the gateway by default. Note: Anthropic's Fable 5 tier requires 30-day retention on every cloud. Details →
- Prompt caching. Explicit cache_control breakpoints; cached reads bill at 0.1x the input price, cache writes carry a 1.25x (5-minute TTL) or 2x (1-hour TTL) surcharge; per-model minimum prefix sizes apply. Cache reads from $0.1/M.
Where Anthropic sits · input price across 60 chat models
- Alibaba $0.05 - $2.5
- ByteDance $0.1 - $0.5
- Google $0.1 - $2
- DeepSeek $0.138 - $1.32
- Tencent $0.14
- OpenAI $0.2 - $30
- MiniMax $0.22 - $0.3
- Moonshot $0.574 - $3
- Anthropic $1 - $10
- Z.ai $1 - $1.4
Each bar spans a vendor’s cheapest to most expensive chat model, so you can see where this one sits against the rest of the catalogue.
Anthropic models & pricing
| Model | Input /M | Output /M | Cache read /M | Context |
|---|---|---|---|---|
| Claude Fable 5invited beta | $10 | $50 | $1 | 200K≈111k words |
| Claude Haiku 4.5 | $1 | $5 | $0.1 | 200K≈150k words |
| Claude Opus 4.5 | $5 | $25 | $0.5 | 200K≈150k words |
| Claude Opus 4.6 | $5 | $25 | $0.5 | 1M≈750k words |
| Claude Opus 4.7 | $5 | $25 | $0.5 | 1M≈555k words |
| Claude Opus 4.8 | $5 | $25 | $0.5 | 1M≈555k words |
| Claude Opus 5 | $5 | $25 | $0.5 | 1M≈555k words |
| Claude Sonnet 4invited beta | $3 | $15 | $0.3 | 200K |
| Claude Sonnet 4.5 | $3 | $15 | $0.3 | 200K≈150k words |
| Claude Sonnet 4.6 | $3 | $15 | $0.3 | 1M≈750k words |
| Claude Sonnet 5 | $2 | $10 | $0.2 | 1M≈555k words |
Provider list prices: no platform markup. Prices refresh live on each model's page. These are official list prices. Logged-in customers may see effective prices including workspace discounts on /console/pricing. Context windows on this page are not directly comparable: this vendor uses more than one tokenizer, so the same token count holds different amounts of text. The word equivalent on each row is the vendor’s own published ratio.
Use Anthropic models in 30 seconds
from openai import OpenAI
client = OpenAI(
base_url="https://synthorai.io/v1",
api_key="sk-syn-...",
)
resp = client.chat.completions.create(
model="claude-opus-4-5",
messages=[{"role": "user", "content": "Summarize this diff"}],
reasoning_effort="medium",
)
print(resp.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://synthorai.io/v1",
apiKey: "sk-syn-...",
});
const resp = await client.chat.completions.create({
model: "claude-opus-4-5",
messages: [{ role: "user", content: "Summarize this diff" }],
reasoning_effort: "medium",
});
console.log(resp.choices[0].message.content);curl https://synthorai.io/v1/chat/completions \
-H "Authorization: Bearer sk-syn-..." \
-H "Content-Type: application/json" \
-d '{
"model": "claude-opus-4-5",
"messages": [{"role": "user", "content": "Hello"}],
"reasoning_effort": "medium"
}'package main
import (
"context"
"fmt"
"github.com/openai/openai-go/v3"
"github.com/openai/openai-go/v3/option"
)
func main() {
client := openai.NewClient(
option.WithBaseURL("https://synthorai.io/v1"),
option.WithAPIKey("sk-syn-..."),
)
resp, _ := client.Chat.Completions.New(context.TODO(), openai.ChatCompletionNewParams{
Model: "claude-opus-4-5",
Messages: []openai.ChatCompletionMessageParamUnion{
openai.UserMessage("Summarize this diff"),
},
ReasoningEffort: openai.ReasoningEffortMedium,
})
fmt.Println(resp.Choices[0].Message.Content)
}import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.chat.completions.*;
import com.openai.models.ReasoningEffort;
OpenAIClient client = OpenAIOkHttpClient.builder()
.baseUrl("https://synthorai.io/v1")
.apiKey("sk-syn-...")
.build();
ChatCompletion resp = client.chat().completions().create(
ChatCompletionCreateParams.builder()
.model("claude-opus-4-5")
.addUserMessage("Summarize this diff")
.reasoningEffort(ReasoningEffort.MEDIUM)
.build());
System.out.println(resp.choices().get(0).message().content().orElse(""));About Anthropic
Anthropic builds the Claude model family, from the frontier Fable/Mythos tier through Opus, Sonnet and Haiku, known for strong reasoning, coding and agentic tool use with large context windows. On Synthorai every Claude model sits behind one OpenAI-compatible endpoint (swap the base_url, keep your code) with prompt-cache pricing passed through at provider rates.
FAQ
How do I get an Anthropic API key without a subscription?
You don't need an Anthropic account: sign up on Synthorai, create one API key, and every Anthropic model on this page works through the OpenAI-compatible endpoint: pay-as-you-go, no subscription (models marked "invited beta" additionally need an approved application).
What does the Anthropic API cost?
Token-billed models start at $1/M input tokens; every model's live per-unit price is in the table above. These are provider list prices with no platform markup.
Can I use my own Anthropic key (BYOK)?
Yes. Bring your own Anthropic key, or your AWS Bedrock / GCP Vertex credentials. Zero markup, free. Your key stays in an encrypted vault and requests using it bill at provider list price.
What is Anthropic's data-retention policy via API?
Zero prompt retention at the gateway by default. Note: Anthropic's Fable 5 tier requires 30-day retention on every cloud.
Other providers