Alibaba: qwen3.5-plus
qwen3.5-plus
Qwen3.5 native vision-language Plus model built on a hybrid architecture that combines linear attention with sparse MoE for higher inference efficiency. Across many benchmarks, the 3.5 series reaches performance comparable to top frontier models, with major gains over the 3 series in both text-only and multimodal tasks. This release is functionally equivalent to the qwen3.5-plus-2026-02-15 snapshot.
Provider: Alibaba|Input types
Output types|Publish Time: None
Group price
Price information for different user groups
Auto Group routing → default
| Group | Billing type | Input Price | Output Price |
|---|---|---|---|
| default | Pay as you go | $0.1100 / M tokens | $0.6600 / M tokens |
Provider
Pricing$0.1100 / M tokens
Input video-
Input audio-
Web Search-
Cache pricing-
Context window-
Max output-
Latency397 ms
Throughput170 TPS
Availability97.50%
Chat
Start a conversation
Type a message below to begin
API call example
Connect quickly using the standard OpenAI-compatible APIPython
1
import openai2
3
client = openai.OpenAI(4
api_key="<YOUR_API_KEY>",5
base_url="http://localhost:3000/v1"6
)7
8
response = client.chat.completions.create(9
model="qwen3.5-plus",10
messages=[11
{"role": "user", "content": "What model are you?"}12
]13
)14
15
print(response.choices[0].message.content)