Requesty
Novita AI logo
Novita AI

qwen/qwen3.5-397b-a17b

The Qwen3.5 series 397B-A17B native vision-language model is based on a hybrid architecture design that integrates linear attention mechanisms with sparse Mixture-of-Experts (MoE), achieving higher inference efficiency. Across a variety of tasks—including language understanding, logical reasoning, code generation, agentic tasks, image understanding, video understanding, and graphical user interface (GUI) interaction—it demonstrates exceptional performance comparable to current top-tier frontier models. Possessing robust code generation and agentic capabilities, it exhibits strong generalization across various agent scenarios.

VisionReasoningTool calling
Try qwen/qwen3.5-397b-a17bCompareDocs

Specifications

Context window262K tokens
Max output66K tokens
API typechat
AddedNov 28, 2025
Model IDnovita/qwen/qwen3.5-397b-a17b
Data retentionYes
Used for trainingUnknown
Provider location🇺🇸 US
Privacy policyNovita AI Privacy Policy→

Benchmarks

Released 2026-02-16
Coding Indexcoding
48.2%

Artificial Analysis Coding Index — a composite of coding evaluations including LiveCodeBench, SciCode and Terminal-Bench.

GPQA Diamondreasoning
89.3%

Graduate-level physics, chemistry & biology questions designed to resist Googling.

Intelligence Indexreasoning
33.7%

Artificial Analysis Intelligence Index — a composite of multiple evaluations measuring overall model capability.

Scores are sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and do not capture every aspect of model quality. Always test on your own workload.

Pricing

Input / 1M
$0.60
Output / 1M
$3.60
Cache write
N/A
Cache read
N/A
Estimated cost
100K input + 10K output$0.0960
1M input + 100K output$0.96
10M input + 1M output$9.60

Requesty charges exactly what the upstream provider charges, no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to novita/qwen/qwen3.5-397b-a17b.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="novita/qwen/qwen3.5-397b-a17b",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other Novita AI models

zai-org/glm-5.1
205K ctx$1.38/1M
Coding 56
xiaomimimo/mimo-v2-pro
1.0M ctx$2.00/1M
xiaomimimo/mimo-v2-flash
262K ctx$0.10/1M
minimax/minimax-m2.7-highspeed
205K ctx$0.60/1M
kwaipilot/kat-coder-pro
256K ctx$0.30/1M
inclusionai/ring-2.6-1t
262K ctx$0.30/1M
Coding 43

Frequently asked questions

How much does qwen/qwen3.5-397b-a17b cost?
qwen/qwen3.5-397b-a17b is priced at $0.60 per million input tokens and $3.60 per million output tokens when accessed via Requesty. Requesty charges exactly what the upstream provider charges, we don't add markup.
What is the context window of qwen/qwen3.5-397b-a17b?
qwen/qwen3.5-397b-a17b has a context window of 262K tokens, with a maximum output of 66K tokens per response. That's roughly 350 words of input you can fit in a single prompt.
How does qwen/qwen3.5-397b-a17b perform on benchmarks?
qwen/qwen3.5-397b-a17b scores 95.6% on τ²-Bench, 89.3% on GPQA Diamond, 48.2% on Coding Index. See the full benchmark chart above for results across MMLU Pro, GPQA Diamond, SWE-Bench Verified, HumanEval, MATH, AIME, MMMU, and LiveBench.
What can qwen/qwen3.5-397b-a17b do?
qwen/qwen3.5-397b-a17b supports vision input, tool calling, extended reasoning. You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use qwen/qwen3.5-397b-a17b with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "novita/qwen/qwen3.5-397b-a17b". The Quickstart above shows Python, JavaScript and cURL snippets.
Can I run qwen/qwen3.5-397b-a17b through Requesty?
Yes. qwen/qwen3.5-397b-a17b runs through Requesty's OpenAI-compatible API, served from Novita AI. You do not host the model yourself: point base_url at Requesty, set the model to "novita/qwen/qwen3.5-397b-a17b", and requests are routed to the upstream provider with automatic failover. The same key gives you 400+ other models too.

Access qwen/qwen3.5-397b-a17b through Requesty

One API key, 400+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

Get started freeAll Novita AI models