Inference APIs
Models/Chat/Qwen3-VL 235B

Qwen3-VL 235B API

Available Vision

Qwen3-VL 235B accepts images and text in the same chat request, in the OpenAI image_url format, as a URL or a base64 data URI. In our tests it read the text in an image exactly, returned it as JSON on request, described an image fetched from a URL, and filled a tool call from an invoice image. It is served at FP8 precision by our second upstream, in the United States.

Try in playground
VisionOCRChatTool callingJSON mode262K context
Context window
262K tokens
Modality
Image + Text → Text
Input
$0.26 / 1M tokens
Output
$1.15 / 1M tokens
Time to first token
1.28 s
Output speed
10.7 tok/s
Model ID
Qwen/Qwen3-VL-235B-A22B-Instruct

Quickstart

You need an API key to call this model. Create a free account or log in.

Set INFERENCE_API_KEY to your key from the API Keys page, then run:

curl https://api.inferenceapis.com/v1/chat/completions \
  -H "Authorization: Bearer $INFERENCE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "Qwen/Qwen3-VL-235B-A22B-Instruct",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello! What can you do?"}
    ]
  }'
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.inferenceapis.com/v1",
    api_key=os.environ["INFERENCE_API_KEY"],
)
response = client.chat.completions.create(
    model="Qwen/Qwen3-VL-235B-A22B-Instruct",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello! What can you do?"},
    ],
)
print(response.choices[0].message.content)
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.inferenceapis.com/v1",
  apiKey: process.env.INFERENCE_API_KEY,
});
const response = await client.chat.completions.create({
  model: "Qwen/Qwen3-VL-235B-A22B-Instruct",
  messages: [
    { role: "system", content: "You are a helpful assistant." },
    { role: "user", content: "Hello! What can you do?" },
  ],
});
console.log(response.choices[0].message.content);

When to use it

A good fit when
  • Reading screenshots, invoices, receipts and scanned pages into text or JSON
  • Agents that need to look at an image and then call a tool
Look elsewhere when
  • Text-only work; the text models here are cheaper

What it costs in practice

WorkloadCost
Chat assistant — 10,000 turns of 800 tokens in, 300 out$5.53
Long prompts or RAG — 10,000 requests of 8,000 in, 500 out$26.55
Generation-heavy — 10,000 requests of 500 in, 2,000 out$24.30

Computed from the live rates below. This model does not reason before answering, so output tokens are the visible answer only. There is no subscription or minimum; new accounts start with free credit.

Endpoint

POST https://api.inferenceapis.com/v1/chat/completions
Headers Authorization: Bearer $INFERENCE_API_KEY
Content-Type: application/json
Compatible OpenAI wire format. /openai/v1/… is accepted too, so Groq-style base URLs work with only the host changed.

Model IDs and aliases

Send any of these as model; they all resolve to this model and bill at its rate.

IDNote
Qwen/Qwen3-VL-235B-A22B-InstructCanonical
qwen3-vlAlias
qwen3-vl-235bAlias

Request body

ParameterTypeDescription
model required string Model ID: Qwen/Qwen3-VL-235B-A22B-Instruct — also accepted: qwen3-vl, qwen3-vl-235b
messages required array Conversation so far. Each item has a role (system, user or assistant) and content.
max_tokens integer Maximum number of tokens to generate.
temperature number Sampling temperature. Higher values give more varied output.
top_p number Nucleus sampling cutoff.
stream boolean Stream the response as server-sent events. The final chunk includes usage.
tools array Tool definitions for function calling (OpenAI format).
response_format object {"type": "json_object"} for JSON mode.
stop string | array Sequences where the model stops generating.
seed integer Seed for more reproducible sampling.

Response

200 OK · application/json
{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "model": "Qwen/Qwen3-VL-235B-A22B-Instruct",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hi! I can answer questions, write and edit text, help with code and more."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": { "prompt_tokens": 28, "completion_tokens": 17, "total_tokens": 45 }
}

Sending images

Put image parts next to the text in a message's content array. A URL and a base64 data URI both work; PNG and JPEG were tested.

Python · read an invoice into JSON
import base64, os
from openai import OpenAI

client = OpenAI(base_url="https://api.inferenceapis.com/v1", api_key=os.environ["INFERENCE_API_KEY"])
b64 = base64.b64encode(open("invoice.png", "rb").read()).decode()

resp = client.chat.completions.create(
    model="Qwen/Qwen3-VL-235B-A22B-Instruct",
    messages=[{"role": "user", "content": [
        {"type": "text", "text": "Return JSON with keys invoice_number and total."},
        {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{b64}"}},   # or a plain https:// URL
    ]}],
    response_format={"type": "json_object"},
    max_tokens=300,
)
print(resp.choices[0].message.content)   # {"invoice_number": "48213", "total": "$1,274.50"}

That output is from our own test image. Images are billed as input tokens, so downscale large ones; text in the image stays readable well below full resolution.

Tested on this endpoint

Tool calling (tools, tool_calls)Works
JSON mode (response_format: json_object)Works
Schema-constrained output (response_format: json_schema)Works
Streaming with usage in the final chunkWorks
System prompt followedWorks

Each row is a live request we sent on September 17, 2026, not a claim copied from a model card. Streamed and parallel tool calls and tool_choice were tested too: full matrix for every model. Request examples are in the API documentation.

Pricing

Input$0.26 / 1M tokens
Output$1.15 / 1M tokens
BillingPay as you go from prepaid credits. No subscription, no daily request or token cap. See all pricing.

Notes

  • It generates slowly here: about 11 tokens per second in our measurement, so a long description takes a while. It suits extraction, where the answer is short, better than long-form writing.
  • Images are billed as input tokens; a 640×240 image came to about 170 tokens and a 1200×630 image to about 770 in our tests. Downscale large images before sending them.
  • The smaller Qwen3-VL 30B is not offered: at our upstream it did not answer requests when we tested it.

Compare Qwen3-VL 235B with…

GPT-OSS 120BDeepSeek V4 FlashDeepSeek V4.1 FlashGLM 5.3 FlashLlama 3.3 70B InstructGLM 5.3DeepSeek V4 ProMiniMax M3Kimi K2.7 CodeGPT-OSS 120B on GroqLlama 3.3 70B on GroqGPT-4o miniGPT-4.1 miniGemini 2.5 FlashDeepSeek V4 Flash (DeepSeek API)

Errors

Errors use the OpenAI envelope: {"error": {"message", "type", "code"}}. 401 missing or invalid key · 402 insufficient_balance · 404 model_not_found · 503 model_unavailable · 502 backend error, safe to retry. Full table with what to retry: API documentation. Errors from other providers: provider error reference.