alibabaCertifiedReleased 09/23/2025

Qwen3 VL 235B A22B Instruct

The Qwen3 series VL models has been comprehensively upgraded in areas such as visual coding and spatial perception. Its visual perception and recognition capabilities have significantly improved, supporting the understanding of ultra-long videos, and its OCR functionality has undergone a major enhancement.

View API reference ↓
Input / output priceFrom$0.20 / $0.88per 1M
24h uptime100.00%
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.inferencebureau.com/v1",
  apiKey: process.env.IB_API_KEY,
});

const res = await client.chat.completions.create({
  model: "alibaba/qwen3-vl-instruct",
  messages: [{ role: "user", content: "Why is the sky blue?" }],
});

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

Providers

Dispatch routes requests across these providers and fails over between them without changing the model. Restrict or order providers with team Directives. Prices are gateway list prices before the platform fee.

ProviderContextMax outputLatencyThroughputInputOutputCacheCapabilitiesZDRNo trainingRegionsStatus
DeepInfra262K262K2.8 s5 tps$0.20$0.88Read $0.11Write —4 featuresNoYesGlobalCertified

Uptime

24 hours

Request success rate per provider.

DeepInfra100.00%

Throughput

last hour

P50 output tokens per second.

DeepInfra5 tps

Latency

last hour

P50 time to first token.

DeepInfra2.8 s

Getting started

Create a key on the API Keys page and set it as IB_API_KEY. See the quickstart.

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.inferencebureau.com/v1",
  apiKey: process.env.IB_API_KEY,
});

const res = await client.chat.completions.create({
  model: "alibaba/qwen3-vl-instruct",
  messages: [{ role: "user", content: "Why is the sky blue?" }],
});

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

Top-level parameters

Standard Chat Completions parameters work as documented. These have model-specific behavior.

ParameterTypeRequiredDescription
modelstringYesModel ID in the form creator/model, e.g. alibaba/qwen3-vl-instruct. Must be allowed by your team's Directives.
max_tokensnumberNoHard cap on generated tokens, and the basis for the balance reserved per request. Qwen3 VL 235B A22B Instruct supports up to 262,144 output tokens.
toolsTool[]NoFunction definitions the model can call. Tool calls are returned in choices[].message.tool_calls.
Supportedmax_tokenstemperaturestoptoolstool_choice

Input limits

InputSourcesLimits
Text—Prompt and response share the 262K-token context window
ImageURL, base64Sent as image_url content parts; counts as input tokens

More models by alibaba

ModelContextInputOutputCacheRelease
alibaba/qwen3.8-max-prime1M$4.00$12.00Read $0.50Write $5.0009/23/2026
alibaba/qwen3.8-omni-flash1M$0.15$0.47Read $0.016Write —09/17/2026
alibaba/qwen3.8-max-0902991K$2.00$6.00Read $0.25Write $2.5009/01/2026
alibaba/qwen3.8-flash991K$0.15$0.47Read $0.016Write $0.2008/26/2026
alibaba/qwen3.8-27b262K$0.060$2.30Read $0.059Write —08/14/2026
alibaba/qwen3.8-2.4t-a95b262K$2.00$6.00Read $0.25Write —08/02/2026
alibaba/qwen3.8-max262K$2.00$6.00Read $0.25Write —08/02/2026
alibaba/qwen3.7-flash991K$0.030$0.13Read $0.0060Write $0.03807/28/2026
alibaba/qwen3.7-plus1M$0.40$1.60Read $0.080Write $0.5006/02/2026
alibaba/qwen3.7-max991K$2.50$7.50Read $0.50Write $3.1305/21/2026