Qwen3.6 35B A3B API

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

Alibaba Cloudテキスト生成128K コンテキストリリース 2026年7月29日ネイティブ推論新着

Qwen3.6 35B A3B について

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.

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

reasoningfunction callingjson modecache

Qwen3.6 35B A3B の仕様

モデルID
qwen3-6-35b-a3b
プロバイダー
Alibaba Cloud
カテゴリ
テキスト生成
リリース
2026年7月29日
コンテキスト長
128K トークン
最大出力
16,384 トークン
入力
テキスト
出力
テキスト
構造化出力
JSON Schema
エンドポイント
POST/v1/chat/completionsPOST/v1/responsesPOST/v1/messagesPOST/v1/completionsPOST/v1beta/models/qwen3-6-35b-a3b:generateContent
代替モデルID
qwen3.6-35b-a3bqwen/qwen3.6-35b-a3bEschaLabs/Qwen3.6-35B-A3B-Escha-W2

Qwen3.6 35B A3B API の料金最大72%お得

EmpirioLabs カタログのライブ従量課金料金です。使った分だけの支払いで、月額最低料金はありません。

タイプ
仕様
料金
入力
per 1M prompt tokens
$0.248$0.07
出力
per 1M generated tokens
$1.485$0.42
Implicit cache read
per 1M cached input tokens
$0.035
Web Search (Linkup)
per call when invoked
$0.013
料金ページ全体で比較

Qwen3.6 35B A3B API の呼び出し方

Qwen3.6 35B A3B は OpenAI 互換の Chat Completions API を提供します。任意の OpenAI SDK を EmpirioLabs API キーで https://api.empiriolabs.ai/v1 に向け、モデル ID qwen3-6-35b-a3b を使ってください。 EmpirioLabs ダッシュボードで API キーを取得してください。

cURL
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."}
    ]
  }'
Python (OpenAI SDK)
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 の完全リファレンス

Qwen3.6 35B A3B API のパラメーター

EmpirioLabs 上の Qwen3.6 35B A3B API が対応するリクエストパラメーターです。フィールドを省略した場合はデフォルト値が適用されます。

パラメータタイプデフォルト範囲 / 値説明
temperaturenumber0.70 から 2Sampling temperature. 0 is deterministic and 2 is maximum randomness.
top_pnumber0.950 から 1Nucleus sampling probability mass. Lower values make outputs more focused.
max_tokensnumber40961 から 16384Maximum output tokens. With thinking on, leave room for the answer as well as the reasoning.
stopstring--Up to 4 strings where the model will stop generating further tokens.
enable_thinkingbooleantrue-Enable the model reasoning channel before final output. Reasoning is returned in reasoning_content and bills as output tokens.
reasoning_effortenummediumnone, low, medium, high, maxReasoning effort. none disables thinking. Any other value enables it.
top_knumber201 から 200Limit sampling to the top K candidate tokens.
min_pnumber00 から 1Minimum probability threshold for token sampling.
presence_penaltynumber0-2 から 2Penalize tokens that have already appeared, increasing topic diversity.
frequency_penaltynumber0-2 から 2Penalize tokens in proportion to how often they have appeared.
seednumber-0 から 2147483647Best-effort determinism. Output is not bit-reproducible across requests because batch composition changes accumulation order.
response_formatenum-text, json_object, json_schemaStructured output. json_object returns valid JSON. json_schema enforces your schema exactly.
web_search_linkupbooleanfalse-Optional web search powered by Linkup. When enabled, recent web sources are retrieved using your latest user message as the query and provided to the model as additional context. Adds $0.013 per call when invoked on top of the model's normal token cost. Disabled by default.

知っておくと便利

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: よくある質問

Qwen3.6 35B A3B API の料金はいくらですか?

EmpirioLabs では Qwen3.6 35B A3B は従量課金です。このページのライブ料金表は常に API の請求額と一致します。

Qwen3.6 35B A3B のコンテキストウィンドウはどれくらいですか?

Qwen3.6 35B A3B は 128K トークンのコンテキストウィンドウに対応しています(1回の応答あたり最大 16,384 出力トークン)。

Qwen3.6 35B A3B API は OpenAI 互換ですか?

はい。Qwen3.6 35B A3B は OpenAI 互換の Chat Completions API を提供しているため、既存の OpenAI SDK は base_urlhttps://api.empiriolabs.ai/v1 に向けてモデル ID を qwen3-6-35b-a3b に設定するだけで動作します。

統合する前にブラウザで Qwen3.6 35B A3B を試せますか?

はい。EmpirioLabs プレイグラウンドでは API と同じパラメーターで Qwen3.6 35B A3B をブラウザ上で実行できるので、コードを書く前にプロンプトを試せます。

Qwen3.6 35B A3B の API キーはどうやって取得しますか?

EmpirioLabs アカウントを作成し、ダッシュボードの API Keysでキーを生成してください。課金は従量制クレジットなので、実行したリクエストの分だけ支払います。

より良いエンドポイントを使う準備はできていますか?

ご自身のモデルを当社のスタックに展開したい方は、ぜひご価格をご覧ください。