Qwen3 235B-A22B

AlibabaMCP Tool CallReasoning
alibaba/qwen3-235b-a22bOpen Weights

Large open Qwen MoE for multilingual reasoning, coding, and tool use

Benchmark Pricing$0.70 / 1M in$2.80 / 1M out
Context: 131,072 tokens•Max Output: 16,384 tokens•Knowledge: 2025-04•Released: 2025-04-28

TECHNICAL SPECIFICATIONS

Context Window131,072131.1K tokens
Max Output16,38416.4K tokens
LicenseOpen SourceSelf-hostable
Input Modalities
Text
Function / Tool CallingSupported (MCP Ready)
Structured OutputsGuaranteed Schema

MODEL CONTEXT PROTOCOL (MCP) READINESS

✓ Qwen3 235B-A22B natively supports OpenAI / Anthropic format tool calling and is fully compatible with MCP servers and agent loops.

MCP Context Window Recommendations

With a context capacity of 131,072 tokens, this model can handle moderately large MCP tool definitions and multi-turn conversation logs comfortably.

STANDARDIZED BENCHMARKS & EVALS

Aider Polyglotpercent correct
59.6
SWE-Bench Proresolve rate
21.41

API CALL INVOCATION EXAMPLE

// Example: Calling Qwen3 235B-A22B via OpenAI SDK
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.ALIBABA_API_KEY,
});

const response = await client.chat.completions.create({
  model: "alibaba/qwen3-235b-a22b",
  messages: [{ role: "user", content: "Hello! Can you inspect MCP tools?" }],
});

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

PRICING BY PROVIDER

15 providers
Alibabaqwen3-235b-a22b
$0.70 / $2.80
Cortecsqwen3-235b-a22b-instruct-2507
$0.20 / $0.60$0.02 cache
NanoGPTqwen/qwen3-235b-a22b-instruct-2507
$0.09 / $0.35$0.02 cache
Scalewayqwen3-235b-a22b-instruct-2507
$0.84 / $2.52
GreenPTqwen3-235b-a22b-instruct-2507
$1.03 / $3.08
302.AIqwen3-235b-a22b-instruct-2507
$0.29 / $1.14
LLM Gatewaynovita/qwen3-235b-a22b-fp8
$0.20 / $0.80
Venice AIqwen3-235b-a22b-instruct-2507
$0.15 / $0.75
Eden AIscaleway/qwen3-235b-a22b-instruct-2507
$0.84 / $2.53
Merge Gatewayqwen/qwen3-235b-a22b-instruct-2507
$0.10 / $0.60$0.06 cache
Hugging FaceQwen/Qwen3-235B-A22B
$0.20 / $0.80
Amazon Bedrockqwen.qwen3-235b-a22b-2507-v1:0
$0.22 / $0.88
Kilo Gatewayqwen/qwen3-235b-a22b-thinking-2507
$0.45 / $3.50
Vertexqwen/qwen3-235b-a22b-instruct-2507-maas
$0.22 / $0.88
DevPass (LLM Gateway)qwen3-235b-a22b-fp8
$0.20 / $0.80

Rates in USD per 1M tokens. Rates synchronized from models.dev public catalog.

MODEL RESOURCES

View Alibaba HubHugging Face Weights

More Models by Alibaba

Alternative and family models from the same lab

View all Alibaba models →
Qwen3 Coder Plus
alibaba/qwen3-coder-plus•Alibaba
1M ctx

Hosted Qwen coder for software agents, repo edits, and long-context code

65.5K out max
In: $1.00 • Out: $5.00
Alibaba
Qwen Turbo
alibaba/qwen-turbo•Alibaba
1M ctx

Efficient Qwen model for fast chat, extraction, and high-volume workloads

16.4K out max
In: $0.05 • Out: $0.20
Alibaba
Qwen3.8 Max
alibaba/qwen3.8-max•Alibaba
1M ctx

2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows

131.1K out maxVisionPDFTerminal-Bench: 86.6 avg@10
In: $2.00 • Out: $6.00
Alibaba