Qwen3-VL Flash is the fast, cost-efficient tier of the Qwen3-VL vision-language series with a 256K-token context.
- Input
- text image video $0.05/M
- Output
- text $0.4/M
- Cache read
- $0.022/M
- Context
- 256K
- vs GPT-4o
- ~99%+ cheaper
Price in context
Where the price sits among 60 comparable models
The bar shows how this model’s price compares with every other model of the same kind on Synthorai. The cheapest and the most expensive are named at each end. These are base rates; batch, region and cache-write discounts are on the pricing page.
Specs & limits
Tokens
| Context window (vendor spec) | 262,144 |
|---|---|
| Max output (vendor spec) | 32,768 |
Prompt caching
| How it caches | automatic + explicit |
|---|---|
| Min prefix | 1,024 |
| Lifetime | explicit: 5m, reset on hit |
| Write cost | 1.25x |
Thinking
| Vendor control | enable_thinking + thinking_budget |
|---|---|
| Accepted values | enable_thinking true · false; thinking_budget in tokens |
| Default | off; enable_thinking defaults to false on the qwen3-vl-plus and qwen3-vl-flash series applied when the request sets nothing |
| Can be turned off | Yes |
| Thinking behaviour | Trace returns in reasoning_content; over the budget the reasoning is truncated and the model answers immediately. Not on the preserve_thinking model list. |
| Parameter | reasoning_effort |
| Values | minimal · low · medium · high the gateway's parameter surface - the vendor mapping above applies |
Model
| Modalities | text + image + video → text |
|---|
- Small-size Qwen3-VL visual-understanding model
- hybrid thinking (off by default)
- video input up to 1 hour / 2GB
- structured output in non-thinking mode only
Use Qwen3 VL Flash in 30 seconds
OpenAI-compatible: swap the base_url, keep your SDK. POST /v1/chat/completions
from openai import OpenAI
client = OpenAI(
base_url="https://synthorai.io/v1",
api_key="sk-syn-...",
)
resp = client.chat.completions.create(
model="qwen3-vl-flash",
messages=[{"role": "user", "content": "Summarize this diff"}],
reasoning_effort="medium",
)
print(resp.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://synthorai.io/v1",
apiKey: "sk-syn-...",
});
const resp = await client.chat.completions.create({
model: "qwen3-vl-flash",
messages: [{ role: "user", content: "Summarize this diff" }],
reasoning_effort: "medium",
});
console.log(resp.choices[0].message.content);curl https://synthorai.io/v1/chat/completions \
-H "Authorization: Bearer sk-syn-..." \
-H "Content-Type: application/json" \
-d '{
"model": "qwen3-vl-flash",
"messages": [{"role": "user", "content": "Hello"}],
"reasoning_effort": "medium"
}'package main
import (
"context"
"fmt"
"github.com/openai/openai-go/v3"
"github.com/openai/openai-go/v3/option"
)
func main() {
client := openai.NewClient(
option.WithBaseURL("https://synthorai.io/v1"),
option.WithAPIKey("sk-syn-..."),
)
resp, _ := client.Chat.Completions.New(context.TODO(), openai.ChatCompletionNewParams{
Model: "qwen3-vl-flash",
Messages: []openai.ChatCompletionMessageParamUnion{
openai.UserMessage("Summarize this diff"),
},
ReasoningEffort: openai.ReasoningEffortMedium,
})
fmt.Println(resp.Choices[0].Message.Content)
}import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.chat.completions.*;
import com.openai.models.ReasoningEffort;
OpenAIClient client = OpenAIOkHttpClient.builder()
.baseUrl("https://synthorai.io/v1")
.apiKey("sk-syn-...")
.build();
ChatCompletion resp = client.chat().completions().create(
ChatCompletionCreateParams.builder()
.model("qwen3-vl-flash")
.addUserMessage("Summarize this diff")
.reasoningEffort(ReasoningEffort.MEDIUM)
.build());
System.out.println(resp.choices().get(0).message().content().orElse(""));About Qwen3 VL Flash
- Per the official Model Studio docs it handles single- and multi-image input, video frame analysis, image captioning, visual question answering, 2D and 3D object localization, and document parsing into HTML or Markdown, along with OCR over receipts, certificates and forms.
- It is a hybrid thinking model: it can reason step by step before answering or respond directly, with thinking disabled by default, switched with enable_thinking and capped with thinking_budget, and the trace returned in a separate reasoning_content field.
- Alibaba calls it the small-size model of the series and names the jobs it was shaped for: security monitoring, store and site inspection rounds, and photographed problem solving.
- That is a sharper answer than "cheaper Plus" to the question of when to pick it.
- The envelope is generous for the tier: up to 32,768 output tokens alongside a large separate reasoning allowance, video accepted up to an hour and 2 GB, and per-image limits governed by a token ceiling rather than a fixed image count.
- Video is sampled by frame extraction you control, with a frames-per-second setting and a frame cap, and pre-extracted frame lists accepted in place of a file.
- Tool calling, batch inference and context caching are supported, with structured output documented for non-thinking mode.
- Weights for the hosted model are not released, though the wider Qwen3-VL family is open.
- Synthorai serves it for multimodal chat via the OpenAI-compatible endpoint.
FAQ
Is the Qwen3 VL Flash API free to try?
Yes: new accounts get 10 trial calls and up to $1 in free credit, no card required. At $0.05/M input tokens, that credit alone covers roughly 2,500 requests of ~8K tokens against Qwen3 VL Flash.
What is Qwen3 VL Flash best at?
2D and 3D object localization; document parsing into HTML or Markdown; hybrid thinking, disabled by default. See the About section for the full picture from the vendor's own release notes.
How much does Qwen3 VL Flash cost?
Qwen3 VL Flash costs $0.05 per million input tokens and $0.4 per million output tokens on Synthorai. That is the provider's list price, with no platform markup. Cached input tokens bill at $0.022/M.
Does Qwen3 VL Flash support prompt caching?
Yes: automatic caching is on by default, with an explicit mode for guaranteed savings. Cached input tokens bill at $0.022/M vs $0.05/M uncached; prompts need a 1,024-token stable prefix to cache (TTL explicit: 5m, reset on hit). Prompt caching guide →
How do I get access to Qwen3 VL Flash?
Point your existing OpenAI SDK at base_url="https://synthorai.io/v1", set model="qwen3-vl-flash", and you're done. One API key covers every model on the gateway.
Related models
Compare
Every value on this page is transcribed from the vendor's own documentation, linked above, and carries the date it was checked. Prices are compared across the catalogue; specification values that vendors define differently are shown with the difference stated rather than charted. Nothing here is measured by us, and nothing is scored.