Qwen3 30B A3B Instruct 2507

Qwen3 30b A3b Instruct is an Alibaba Qwen model for multilingual generation, reasoning, coding, and multimodal tasks.

Input / output price: $0.10 / $0.30 per 1M
Context: 262K tokens
Released: Jul 30, 2025

Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "qwen3-30b-a3b-instruct-2507".

from openai import OpenAI

client = OpenAI(

    base_url="https://api.respan.ai/api/",

    api_key="YOUR_RESPAN_API_KEY",

)

response = client.chat.completions.create(

    model="qwen3-30b-a3b-instruct-2507",

    messages=[{"role": "user", "content": "Hello!"}],

)

print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({

  baseURL: "https://api.respan.ai/api/",

  apiKey: "YOUR_RESPAN_API_KEY",

});

const response = await client.chat.completions.create({

  model: "qwen3-30b-a3b-instruct-2507",

  messages: [{ role: "user", content: "Hello!" }],

});

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

curl https://api.respan.ai/api/chat/completions \

  -H "Authorization: Bearer YOUR_RESPAN_API_KEY" \

  -H "Content-Type: application/json" \

  -d '{

    "model": "qwen3-30b-a3b-instruct-2507",

    "messages": [{"role": "user", "content": "Hello!"}]

}'

Read docs

Providers

2 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.

Provider	Context	Input	Output	Cache read	Capabilities
Nebius`nebius/Qwen/Qwen3-30B-A3B-Instruct-2507`	262K	$0.10/M	$0.30/M	$0.10/M
OpenRouter`openrouter/qwen/qwen3-30b-a3b-instruct-2507`	262K	$0.10/M	$0.30/M	$0.10/M

Is a provider down? OpenRouter status

Pricing

Averaged across 2 providers, in USD per 1M tokens after gateway discounts.

Average input: $0.10 per 1M
Average output: $0.30 per 1M
Average cache read: $0.10 per 1M

Cost by volume

Total cost from 0 to 1T tokens.

$0.10/MNebius and OpenRouter

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

15tokens/sacross providers

Faster than 21% of the other 471 models on the gateway

Response latency

Lower is better

2.6sacross providers

Faster than 47% of the other 496 models on the gateway

Benchmarks

The Respan Index plus 19 published scores from 3 sources, each linked to where it was published.

13.7/ 100

Respan Index#153 of 162

162 ranked models. The shaded band is this model's 95% interval, 10.8 to 20.2. Llama 4 Maverick and gpt-oss-20b fall inside it, so the gap is within the margin of error.

153Qwen3 30B-A3B Instruct 250713.7
152Llama 4 Maverick14.3
154gpt-oss-20b13.6

Published scores

Benchmark	Score	Source	Published
MultiPL-Ehuggingface.co	83.8%	huggingface.co	Jul 30, 2025
LiveCodeBench v6 (25.02-25.05)huggingface.co	43.2%	huggingface.co	Jul 30, 2025
Aider-Polyglothuggingface.co	35.6%	huggingface.co	Jul 30, 2025

Benchmark	Score	Source	Published
BFCL-v3huggingface.co	65.1%	huggingface.co	Jul 30, 2025
TAU2-Retailhuggingface.co	57%	huggingface.co	Jul 30, 2025

Benchmark	Score	Source	Published
ZebraLogichuggingface.co	90%	huggingface.co	Jul 30, 2025
GPQAhuggingface.co	70.4%	huggingface.co	Jul 30, 2025
LiveBench 20241125huggingface.co	69%	huggingface.co	Jul 30, 2025
GPQA DiamondEpoch AI	55.6%	Epoch AI	Oct 5, 2026
SuperGPQAhuggingface.co	53.4%	huggingface.co	Jul 30, 2025
Chess PuzzlesEpoch AI	2%	Epoch AI	Oct 5, 2026

Benchmark	Score	Source	Published
OTIS Mock AIME 2024-2025Epoch AI	62.2%	Epoch AI	Oct 5, 2026
AIME25huggingface.co	61.3%	huggingface.co	Jul 30, 2025
HMMT25huggingface.co	43%	huggingface.co	Jul 30, 2025

Benchmark	Score	Source	Published
MMLU-Reduxhuggingface.co	89.3%	huggingface.co	Jul 30, 2025
IFEvalhuggingface.co	84.7%	huggingface.co	Jul 30, 2025
MMLU-Prohuggingface.co	78.4%	huggingface.co	Jul 30, 2025

Benchmark	Score	Source	Published
Arena-Hard v2huggingface.co	69%	huggingface.co	Jul 30, 2025
LMArena EloLMArena	1382 Elo · #129 of 163	LMArena	Oct 28, 2026

How the Respan Index works

How Qwen compares: DeepSeek vs Qwen, Claude vs Qwen, Gemini vs Qwen

Apps

Use Qwen3 30B A3B Instruct 2507 in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

Qwen3 30B A3B Instruct 2507 FAQ

Short answers, built from the data on this page.

Qwen3 30B A3B Instruct 2507 is a model from Qwen, released on Jul 30, 2025. Qwen3 30b A3b Instruct is an Alibaba Qwen model for multilingual generation, reasoning, coding, and multimodal tasks.

On the Respan gateway, Qwen3 30B A3B Instruct 2507 costs $0.10 per 1M input tokens and $0.30 per 1M output tokens on every provider, after gateway discounts.

Qwen3 30B A3B Instruct 2507 is available through 2 providers: Nebius and OpenRouter.

Qwen3 30B A3B Instruct 2507 accepts up to 262K tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://api.respan.ai/api/ with your Respan API key and set the model to "qwen3-30b-a3b-instruct-2507". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Qwen3 30B A3B Instruct 2507 scores 13.7 out of 100 on the Respan Index v1.1, ranking #153 of 162 models. By category: Coding 1.6 (estimated), Agents 6.5 (estimated), Reasoning 18.1, Knowledge & language 7.9, and Human preference 42.1. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

Qwen3 30B A3B Instruct 2507 has 19 published benchmark scores, including MultiPL-E 83.8%, BFCL-v3 65.1%, and ZebraLogic 90%. Each score links to where it was published; the data is collected by Respan.