Updated

Streaming an abliterated LLM API

Set stream to true and the API sends server-sent events as the model writes. Thinking arrives in delta.reasoning_content, the answer in delta.content, and the stream ends with [DONE].

https://api.silkcompute.com
stream = client.chat.completions.create(
    model="deepseek-v4.1-flash-1017-uncensored",
    messages=[{"role": "user", "content": "Write a commit message for a typo fix."}],
    stream=True,
    stream_options={"include_usage": True},
)

for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)
    if chunk.usage:
        print("\n", chunk.usage)

Example events

data: {"choices":[{"index":0,"delta":{"reasoning_content":"Short, imperative."}}]}

data: {"choices":[{"index":0,"delta":{"content":"Fix typo"}}]}

data: {"choices":[{"index":0,"delta":{"content":" in README"}}]}

data: {"choices":[],"usage":{"prompt_tokens":18,"completion_tokens":40}}

data: [DONE]
const stream = await client.chat.completions.create({
  model: 'deepseek-v4.1-flash-1017-uncensored',
  messages: [{ role: 'user', content: 'Write a commit message for a typo fix.' }],
  stream: true,
  stream_options: { include_usage: true },
})

for await (const chunk of stream) {
  const text = chunk.choices[0]?.delta?.content
  if (text) process.stdout.write(text)
  if (chunk.usage) console.log('\n', chunk.usage)
}

Example events

data: {"choices":[{"index":0,"delta":{"reasoning_content":"Short, imperative."}}]}

data: {"choices":[{"index":0,"delta":{"content":"Fix typo"}}]}

data: {"choices":[{"index":0,"delta":{"content":" in README"}}]}

data: {"choices":[],"usage":{"prompt_tokens":18,"completion_tokens":40}}

data: [DONE]
curl -N https://api.silkcompute.com/chat/completions \
  -H "Authorization: Bearer $SILK_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4.1-flash-1017-uncensored",
    "stream": true,
    "stream_options": {"include_usage": true},
    "messages": [
      {"role": "user", "content": "Write a commit message for a typo fix."}
    ]
  }'

Example events

data: {"choices":[{"index":0,"delta":{"reasoning_content":"Short, imperative."}}]}

data: {"choices":[{"index":0,"delta":{"content":"Fix typo"}}]}

data: {"choices":[{"index":0,"delta":{"content":" in README"}}]}

data: {"choices":[],"usage":{"prompt_tokens":18,"completion_tokens":40}}

data: [DONE]
stream := client.Chat.Completions.NewStreaming(context.TODO(), openai.ChatCompletionNewParams{
	Model: "deepseek-v4.1-flash-1017-uncensored",
	Messages: []openai.ChatCompletionMessageParamUnion{
		openai.UserMessage("Write a commit message for a typo fix."),
	},
})
defer stream.Close()

for stream.Next() {
	chunk := stream.Current()
	if len(chunk.Choices) > 0 {
		fmt.Print(chunk.Choices[0].Delta.Content)
	}
}
if err := stream.Err(); err != nil {
	panic(err)
}

Example events

data: {"choices":[{"index":0,"delta":{"reasoning_content":"Short, imperative."}}]}

data: {"choices":[{"index":0,"delta":{"content":"Fix typo"}}]}

data: {"choices":[{"index":0,"delta":{"content":" in README"}}]}

data: {"choices":[],"usage":{"prompt_tokens":18,"completion_tokens":40}}

data: [DONE]

How it works

OrderThinking, then the answerThinking pieces arrive in delta.reasoning_content before the answer starts in delta.content.
UsageCounted in the last chunkSet stream_options.include_usage to get token counts at the end.
FormatsEvery format streamsMessages and Responses stream too, with stream: true and their own event shapes.

What each chunk carries

FieldHolds
choices[0].delta.reasoning_contentA piece of the thinking
choices[0].delta.contentA piece of the answer
usageToken counts, in the last chunk with include_usage
data: [DONE]End of stream