> ## Documentation Index
> Fetch the complete documentation index at: https://phaseo.app/docs/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Phaseo Gateway base URL: https://api.phaseo.app/v1. Send `Authorization: Bearer $PHASEO_API_KEY` with every request.
> Model IDs look like `provider/model`, for example `openai/gpt-6-astra`. Do not guess IDs: list them with `GET /v1/models` or check https://phaseo.app/models. `phaseo/free` routes to an available free model and needs no wallet balance.
> For new text integrations prefer `POST /v1/responses`. `POST /v1/chat/completions` (OpenAI format) and `POST /v1/messages` (Anthropic format) suit existing code.
> The full API contract is https://phaseo.app/docs/openapi/v1/openapi.yaml. Official SDKs: `@phaseo/sdk` (TypeScript) and `phaseo` (Python), plus Go, C#, PHP, Ruby and Rust.
> Search these docs over MCP at https://phaseo.app/docs/mcp, and install Phaseo skills with `npx skills add https://phaseo.app/docs`. If a page is wrong or out of date, report it with the MCP `submit_feedback` tool.
> Manage Phaseo accounts with the `phaseo` CLI (`npm install -g @phaseo/cli`, then `phaseo login` or `phaseo login --device-code`), the `/v1` Management API or the MCP server at `https://mcp.phaseo.app/mcp?access=manage`. Do not send users to the dashboard for account tasks. A person is still needed for sign-up, approving a sign-in, paying the first top-up on the checkout link that `phaseo credits add` returns when no card is saved, confirming each saved-card top-up from the emailed link, and account security (password, MFA, passkeys). Create API keys with `phaseo keys create --name <name> --json`; never guess one. Credit top-ups need `phaseo login --scopes billing:write` first. Task map: https://phaseo.app/docs/v1/developers/manage-from-an-agent.md.

# Chat Completions

> Rufe /chat/completions aus jedem Phaseo-SDK auf.

Rufe die Chat-Methode des SDK auf, um `POST /v1/chat/completions` zu senden und eine vollständige Antwort zu erhalten. Für neue Integrationen bietet die [Responses API](https://phaseo.app/docs/de/v1/sdk-reference/client/responses.md) eine einheitliche Anfrageform für Text, multimodale Eingaben, Tools und reichhaltigere Ausgabeelemente.

```typescript TypeScript theme={null}
const chat = await client.generateText({
  model: "openai/gpt-6-astra",
  messages: [
    { role: "system", content: "You are helpful." },
    { role: "user", content: "Write a limerick about lighthouses." },
  ],
});

console.log(chat.choices[0].message.content);
```

```python Python theme={null}
chat = client.generate_text({
    "model": "openai/gpt-6-astra",
    "messages": [
        {"role": "system", "content": "You are helpful."},
        {"role": "user", "content": "Write a limerick about lighthouses."},
    ],
})

print(chat["choices"][0]["message"]["content"])
```

```go Go theme={null}
import (
  "context"
  "fmt"

  phaseo "github.com/phaseoteam/Phaseo/packages/sdk/sdk-go/v3"
)

client, err := phaseo.NewFromEnv()
if err != nil {
  panic(err)
}

resp, err := client.GenerateText(context.Background(), phaseo.ChatCompletionsRequest{
  Model: "openai/gpt-6-astra",
  Messages: []map[string]interface{}{
    {"role": "user", "content": "Write a one-line haiku."},
  },
})
if err != nil {
  panic(err)
}
fmt.Println(resp["choices"])
```

```csharp C# theme={null}
using PhaseoSdk;

var client = new Phaseo();

var response = await client.GenerateText(new Dictionary<string, object>
{
    ["model"] = "openai/gpt-6-astra",
    ["messages"] = new[]
    {
        new Dictionary<string, object>
        {
            ["role"] = "user",
            ["content"] = "Write a limerick about lighthouses.",
        },
    },
});
```

```php PHP theme={null}
<?php
require 'vendor/autoload.php';

use Phaseo\Sdk\Phaseo;

$client = new Phaseo();

$response = $client->createChatCompletion([
    'model' => 'openai/gpt-6-astra',
    'messages' => [[
        'role' => 'user',
        'content' => 'Write a limerick about lighthouses.',
    ]],
]);

echo $response['choices'][0]['message']['content'];
```

```ruby Ruby theme={null}
require 'phaseo_sdk'

client = PhaseoSdk::Phaseo.new

response = client.create_chat_completion(
  model: 'openai/gpt-6-astra',
  messages: [{ role: 'user', content: 'Write a limerick about lighthouses.' }]
)
```

```java Java theme={null}
import app.phaseo.sdk.Phaseo;
import java.util.List;
import java.util.Map;

Phaseo client = new Phaseo(System.getenv("PHASEO_API_KEY"));

var response = client.createChatCompletion(Map.of(
    "model", "openai/gpt-6-astra",
    "messages", List.of(
        Map.of("role", "user", "content", "Write a one-line haiku.")
    )
));
System.out.println(response.path("choices").path(0).path("message").path("content").asText());
```

```rust Rust theme={null}
use phaseo::Phaseo;
use serde_json::json;

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let client = Phaseo::from_env()?;
    let response = client.chat_completions(&json!({
        "model": "openai/gpt-6-astra",
        "messages": [
            {
                "role": "system",
                "content": "Answer concisely."
            },
            {
                "role": "user",
                "content": "What is provider fallback?"
            }
        ]
    }))?;

    let text = response
        .body
        .pointer("/choices/0/message/content")
        .and_then(|value| value.as_str())
        .unwrap_or("");

    println!("{text}");
    Ok(())
}
```

| Sprache | Methode | Rückgabe |
| - | - | - |
| TypeScript | `client.generateText(request)`; `client.chat.completions.create(request)` macht dasselbe und wechselt zu Streaming, wenn `stream` true ist | `ChatCompletionsResponse` |
| Python | `client.generate_text(request)`; außerdem `client.chat.completions.create(request)` | `APIResponse`-Dict |
| Go | `client.GenerateText(ctx, req)`, Alias `CreateChatCompletion`; nimmt eine `phaseo.ChatCompletionsRequest` entgegen | `map[string]interface{}` |
| C# | `client.GenerateText(request)`, Alias `CreateChatCompletion` | `Dictionary<string, object>?` |
| PHP | `$client->generateText($request)`, Alias `createChatCompletion()` | Array |
| Ruby | `client.generate_text(...)`, Alias `create_chat_completion` | Hash |
| Java | `client.createChatCompletion(request)`; nimmt eine `Map` oder einen Jackson-`JsonNode` entgegen | Jackson-`JsonNode` |
| Rust | `client.chat_completions(&request)` | `PhaseoResponse`; das JSON steht in `body` |

Jeder Rückgabewert hat die Form von Chat Completions: Die Antwort steht in `choices[0].message.content`, daneben `finish_reason` und `usage`.

## Antwort streamen

```typescript TypeScript theme={null}
for await (const chunk of client.streamChat({
  model: "openai/gpt-6-astra",
  messages: [{ role: "user", content: "Stream a story" }],
})) {
  if (chunk.text) process.stdout.write(chunk.text);
}
```

```python Python theme={null}
for chunk in client.stream_chat({
    "model": "openai/gpt-6-astra",
    "messages": [{"role": "user", "content": "Stream a story"}],
}):
    if chunk.get("text"):
        print(chunk["text"], end="", flush=True)
```

```java Java theme={null}
import app.phaseo.sdk.Phaseo;
import com.fasterxml.jackson.databind.ObjectMapper;
import java.util.List;
import java.util.Map;

Phaseo client = new Phaseo(System.getenv("PHASEO_API_KEY"));
var payload = Map.of(
    "model", "openai/gpt-6-astra",
    "messages", List.of(Map.of("role", "user", "content", "Hello")),
    "stream", true
);
try (var lines = client.rawClient().requestLines(
    "POST", "/chat/completions", null, null,
    new ObjectMapper().writeValueAsString(payload)
)) {
    lines.forEach(System.out::println);
}
```

| Sprache | So streamst du |
| - | - |
| TypeScript | `client.streamChat(request)` liefert geparste Chunks; `client.streamText()` liefert die rohen SSE-Zeilen. |
| Python | `client.stream_chat(request)` liefert geparste Chunks; `client.stream_text()` liefert die rohen SSE-Zeilen. |
| Java | `createChatCompletion` gibt einen einzigen JSON-Body zurück. Setze `stream: true` und sende die Anfrage mit `rawClient().requestLines`, das SSE-Zeilen zurückgibt, sobald sie eintreffen; schließe es mit try-with-resources. |
| Go, C#, PHP, Ruby, Rust | Kein Streaming-Helfer für Chat Completions; lass `stream` ungesetzt. In C#, PHP und Ruby hat [Responses](https://phaseo.app/docs/de/v1/sdk-reference/client/responses.md#stream-the-response) eine Streaming-Methode. |

Jeder geparste Chunk (TypeScript und Python) behält die ursprünglichen Felder (`choices`, `id`, …) und ergänzt `text` (den Inhalt, den dieser Chunk hinzufügt), `usage` und Reasoning-Tokens (`reasoningTokens` / `reasoning_tokens`).

## Parameter

Die Feldnamen sind in jedem SDK die Felder des JSON-Bodys. Die typisierte `phaseo.ChatCompletionsRequest` in Go verwendet Struct-Felder in PascalCase (`Model`, `Messages`).

| Gruppe | Felder |
| - | - |
| Erforderlich | `model`; `messages` mit den Rollen `system`, `developer`, `user`, `assistant` oder `tool` und Inhalt als String oder Parts |
| Sampling | `temperature` (0–2), `top_p` (0–1), `top_k` (-1 oder mehr), `min_p` (0–1), `seed`, sofern das Modell sie unterstützt |
| Länge und Penalties | `max_tokens` oder `max_completion_tokens`, `presence_penalty` und `frequency_penalty` (-2 bis 2), `repetition_penalty` (über 0), `stop` (String oder Array) |
| Tools | `tools`, `tool_choice`, `max_tool_calls`, `parallel_tool_calls` |
| Reasoning | `reasoning`-Objekt oder `reasoning_effort` |
| Log-Wahrscheinlichkeiten | `logprobs`, `top_logprobs` (0–20) |
| Ausgabe | `response_format`, `n` (1–16), `stream`, `service_tier`, `metadata` (String-Werte) |
| Gateway | `provider`-Routing, `usage` (Nutzung einschließen), `meta` (den Meta-Block des Gateways einschließen) |

[Parameter](https://phaseo.app/docs/de/v1/api-reference/parameters.md) erklärt jedes Feld, und die [Chat-Completions-API-Referenz](https://phaseo.app/docs/de/v1/api-reference/endpoint/chat-completions.md) enthält das vollständige Schema.

## Antwort

`ChatCompletionsResponse`:

```json theme={null}
{
  "id": "chatcmpl-123",
  "object": "chat.completion",
  "created": 1677652288,
  "model": "openai/gpt-6-astra",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello there, how may I assist you today?"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 9,
    "completion_tokens": 12,
    "total_tokens": 21
  }
}
```

Mit `stream: true` sendet der Endpunkt SSE-Frames, die mit `data: [DONE]` enden:

```
data: {"id":"chatcmpl-123","object":"chat.completion.chunk","created":1677652288,"model":"openai/gpt-6-astra","choices":[{"index":0,"delta":{"role":"assistant","content":""},"finish_reason":null}]}

data: {"id":"chatcmpl-123","object":"chat.completion.chunk","created":1677652288,"model":"openai/gpt-6-astra","choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}

data: {"id":"chatcmpl-123","object":"chat.completion.chunk","created":1677652288,"model":"openai/gpt-6-astra","choices":[{"index":0,"delta":{},"finish_reason":"stop"}],"usage":{"prompt_tokens":9,"completion_tokens":12,"total_tokens":21}}

data: [DONE]
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.