Last refreshed 2026-10-27. Next refresh: weekly.
Why use Qwen3.5-397B-A17B on DeepInfra?
DeepInfra offers Qwen3.5-397B-A17B with pay-as-you-go pricing at $0.45/1M input tokens. DeepInfra is a cloud inference platform offering cost-effective access to open-source AI models.
Compare Qwen3.5-397B-A17B across 5 providers to find the best fit for your use caseInput / 1M
$0.45
Output / 1M
$3.00
Cache
read $0.22
Batch
Not sourced
Setup recipe
Python + curlInstall
pip install openaiAuth
export DEEPINFRA_API_KEY=...Call
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["DEEPINFRA_API_KEY"],Model ID
Qwen/Qwen3.5-397B-A17BRequest example
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["DEEPINFRA_API_KEY"],
base_url="https://api.deepinfra.com/v1/openai"
)
response = client.chat.completions.create(
model="Qwen/Qwen3.5-397B-A17B",
messages=[{"role": "user", "content": "Hello"}]
)
print(response.choices[0].message.content)Gotchas
- Use provider model ID "Qwen/Qwen3.5-397B-A17B", not the LLMReference slug "qwen3.5-397b-a17b".
- DeepInfra uses "organization/model-name" format, e.g. "meta-llama/Meta-Llama-3-8B-Instruct" or "mistralai/Mistral-7B-Instruct-v0.3". See the DeepInfra model catalog for exact IDs.
- The examples expect DEEPINFRA_API_KEY; rename it only if your application config maps the new variable.
Compare Qwen3.5-397B-A17B Across Providers
| Provider | Input (per 1M) | Output (per 1M) |
|---|---|---|
| OpenRouter | $0.50 | $3.60 |
| Together AI | $0.60 | $3.60 |
| Alibaba Cloud PAI-EAS | $0.60 | $3.60 |
| Novita AI | $0.60 | $3.60 |
| DeepInfra | $0.45 | $3.00 |
Pricing
| Type | Price (per 1M) |
|---|---|
| Input tokens | $0.45 |
| Output tokens | $3.00 |
Capabilities
VisionMultimodalReasoningJSON / Tool useStructured Outputs
About Qwen3.5-397B-A17B
Alibaba's largest Qwen3.5 model, featuring a Mixture-of-Experts architecture with 397B total parameters and 17B active per token (using 512 total experts with 10 routed + 1 shared active). Supports 201 languages with a native 262K token context window extensible to 1M tokens via YaRN. Includes a thinking/reasoning mode, tool calling with MCP integration, and unified vision-language capabilities through early fusion training.
Get Started
Model Specs
Released2026-02-16
Parameters397B (17B active)
Context256K
ArchitectureMixture of Experts