Qwen3.6 35B A3B API

Qwen3.6 35B A3B 소개

Qwen3.6 35B A3B is a 256-expert mixture-of-experts reasoning model with 128K context, function tools, and strict structured JSON output.

Text-only. This build does not accept image or video input, unlike the base Qwen3.6 35B A3B. Weights Served from the 2-bit eschamoe W2 build published by Escha Labs (eschalabs.com) as EschaLabs/Qwen3.6-35B-A3B-Escha-W2 on Hugging Face, under Apache-2.0. The experts are quantized to 2 bits, mixed per projection (gate_up_proj at 2-bit and down_proj at 3-bit), the dense layers are int8, and the KV cache is FP16. Escha Labs publishes the quality comparison against an FP8 baseline of the same model: parity or better on math, graduate science, tool use and long context, about 2 percent lower on broad knowledge, and about 7 percent lower on long-horizon code generation, which is the one clear gap. See the model card for the full benchmark table and protocol. Behavior Supports streaming, function tools, structured JSON output including strict schemas, and thinking mode on by default. Set enable_thinking=false for direct answers. With thinking on, the reasoning arrives in reasoning_content and the answer in content, so read both. A low max_tokens with thinking on can be spent entirely on reasoning, so allow room for the answer. Caching Automatic prefix cache reads are billed at the cached-input rate when reported. Explicit cache controls are not supported. Cancelling a streaming request mid-generation bills only the tokens produced up to that point.

다른 이름 Alibaba Cloud Qwen3.6 35B A3B, Qwen3.6-35B-A3B, qwen3-6-35b-a3b

Qwen3.6 35B A3B 사양

모델 ID

qwen3-6-35b-a3b

제공자

Alibaba Cloud

카테고리

텍스트 생성

출시

2026년 7월 29일

컨텍스트 창

128K 토큰

최대 출력

16,384 토큰

입력

텍스트

출력

텍스트

구조화된 출력

JSON Schema

엔드포인트

POST/v1/chat/completionsPOST/v1/responsesPOST/v1/messagesPOST/v1/completionsPOST/v1beta/models/qwen3-6-35b-a3b:generateContent

대체 모델 ID

qwen3.6-35b-a3bqwen/qwen3.6-35b-a3bEschaLabs/Qwen3.6-35B-A3B-Escha-W2

Qwen3.6 35B A3B API 요금최대 72% 절약

EmpirioLabs 카탈로그의 실시간 종량제 요금입니다. 사용한 만큼만 결제하며 월 최소 요금이 없습니다.

Qwen3.6 35B A3B API 호출 방법

Qwen3.6 35B A3B은(는) OpenAI 호환 Chat Completions API를 제공합니다. 아무 OpenAI SDK나 EmpirioLabs API 키와 함께 https://api.empiriolabs.ai/v1로 지정하고 모델 ID qwen3-6-35b-a3b를 사용하세요. EmpirioLabs 대시보드에서 API 키를 발급받으세요.

curl https://api.empiriolabs.ai/v1/chat/completions \
  -H "Authorization: Bearer $EMPIRIOLABS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3-6-35b-a3b",
    "messages": [
      {"role": "user", "content": "Write a haiku about the ocean."}
    ]
  }'

from openai import OpenAI

client = OpenAI(
    base_url="https://api.empiriolabs.ai/v1",
    api_key="YOUR_EMPIRIOLABS_API_KEY",
)

response = client.chat.completions.create(
    model="qwen3-6-35b-a3b",
    messages=[{"role": "user", "content": "Write a haiku about the ocean."}],
)
print(response.choices[0].message.content)

Qwen3.6 35B A3B API 파라미터

EmpirioLabs에서 Qwen3.6 35B A3B API가 지원하는 요청 파라미터입니다. 필드를 생략하면 기본값이 적용됩니다.

알아두면 좋은 점

Text-only. This build does not accept image or video input, unlike the base Qwen3.6 35B A3B.

Weights

Served from the 2-bit eschamoe W2 build published by Escha Labs (eschalabs.com) as EschaLabs/Qwen3.6-35B-A3B-Escha-W2 on Hugging Face, under Apache-2.0. The experts are quantized to 2 bits, mixed per projection (gate_up_proj at 2-bit and down_proj at 3-bit), the dense layers are int8, and the KV cache is FP16. Escha Labs publishes the quality comparison against an FP8 baseline of the same model: parity or better on math, graduate science, tool use and long context, about 2 percent lower on broad knowledge, and about 7 percent lower on long-horizon code generation, which is the one clear gap. See the model card for the full benchmark table and protocol.

Behavior

Supports streaming, function tools, structured JSON output including strict schemas, and thinking mode on by default. Set enable_thinking=false for direct answers. With thinking on, the reasoning arrives in reasoning_content and the answer in content, so read both. A low max_tokens with thinking on can be spent entirely on reasoning, so allow room for the answer.

Caching

Automatic prefix cache reads are billed at the cached-input rate when reported. Explicit cache controls are not supported. Cancelling a streaming request mid-generation bills only the tokens produced up to that point.

Qwen3.6 35B A3B API: 자주 묻는 질문

Qwen3.6 35B A3B API 비용은 얼마인가요?

EmpirioLabs에서 Qwen3.6 35B A3B은(는) 종량제로 청구됩니다. 이 페이지의 실시간 요금표는 항상 API 청구 금액과 일치합니다.

Qwen3.6 35B A3B의 컨텍스트 윈도우는 얼마인가요?

Qwen3.6 35B A3B은(는) 128K 토큰 컨텍스트 윈도우를 지원합니다 (응답당 최대 16,384 출력 토큰).

Qwen3.6 35B A3B API는 OpenAI와 호환되나요?

네. Qwen3.6 35B A3B은(는) OpenAI 호환 Chat Completions API를 제공하므로, 기존 OpenAI SDK에서 base_url을 https://api.empiriolabs.ai/v1로 지정하고 모델 ID를 qwen3-6-35b-a3b로 설정하면 바로 동작합니다.

통합하기 전에 브라우저에서 Qwen3.6 35B A3B을(를) 사용해 볼 수 있나요?

네. EmpirioLabs 플레이그라운드에서 API와 동일한 파라미터로 Qwen3.6 35B A3B을(를) 브라우저에서 실행하므로 코드를 작성하기 전에 프롬프트를 테스트할 수 있습니다.

Qwen3.6 35B A3B API 키는 어떻게 발급받나요?

EmpirioLabs 계정을 만든 다음 대시보드의 API Keys에서 키를 생성하세요. 요금은 종량제 크레딧이라 실행한 요청에 대해서만 결제합니다.