Alibaba Qwen logo

Alibaba Qwen: Qwen3.5 397B-A17B

Large open Qwen multimodal MoE for visual agents and long technical tasks

Source-linked qwen Open weights Released 2026-02-15
API record Report
InputT
OutputT
Input price$0.6/M
Output price$3.6/M
Context262.144K
Max output65.536K
Providers19
Inference availability

Providers

Provider-specific identifiers, limits, and listed prices per million tokens. Every row links back to the provider's own documentation.

Report a price
Providers offering Qwen3.5 397B-A17B
ProviderProvider model IDContextMax outputInputOutputCache readCapabilitiesDocs
UnoRouter qwen3.5-397b-a17b:free 262.144K 65.536K ReasoningToolsJSON Docs ↗
EmpirioLabs AI qwen3-5-397b-a17b 256K 64K $0.172 $1.032 $0.172 ReasoningToolsJSON Docs ↗
OrcaRouter qwen/qwen3.5-397b-a17b 262.144K 65.536K $0.172 $1.032 ReasoningToolsJSON Docs ↗
Alibaba (China) qwen3.5-397b-a17b 262.144K 65.536K $0.172 $1.032 ReasoningTools Docs ↗
Vultr Qwen/Qwen3.5-397B-A17B 262.144K 65.536K $0.3 $2 ReasoningToolsJSON Docs ↗
CrofAI qwen3.5-397b-a17b 262.144K 262.144K $0.35 $1.75 $0.07 ReasoningToolsJSON Docs ↗
Kilo Gateway qwen/qwen3.5-397b-a17b 262.144K 235.929K $0.39 $2.34 ReasoningToolsJSON Docs ↗
SiliconFlow Qwen/Qwen3.5-397B-A17B 262.144K 262.144K $0.39 $2.34 ToolsJSON Docs ↗
TokenGo qwen/qwen3.5-397b-a17b 262.144K 65.536K $0.4 $2.65 $0.2 ReasoningToolsJSON Docs ↗
Ofox bailian/qwen3.5-397b-a17b 256K 64K $0.55 $3.5 $0.55 ReasoningToolsJSON Docs ↗
Ofox qwen/qwen3.5-397b-a17b 256K 64K $0.55 $3.5 $0.55 ReasoningToolsJSON Docs ↗
OpenRouter qwen/qwen3.5-397b-a17b 262.144K 235.929K $0.55 $3.5 $0.225 ReasoningToolsJSON Docs ↗
Jalapeno Cloud Qwen3.5-397B-A17B 262.144K 65.536K $0.6 $3.6 ReasoningToolsJSON Docs ↗
LLMTR qwen/qwen3.5-397b-a17b 256K 65.536K $0.6 $3.6 ReasoningToolsJSON Docs ↗
Alibaba qwen3.5-397b-a17b 262.144K 65.536K $0.6 $3.6 ReasoningToolsJSON Docs ↗
DevPass (LLM Gateway) qwen35-397b-a17b 262.144K 65.536K $0.6 $3.6 ReasoningToolsJSON Docs ↗
Cloudflare AI Gateway alibaba/qwen3.5-397b-a17b 262.144K 65.536K $0.6 $3.6 ReasoningToolsJSON Docs ↗
Cortecs qwen3.5-397b-a17b 262K 262K $0.668 $4.01 ReasoningTools Docs ↗
GreenPT qwen3.5-397b-a17b 262.144K 16.384K $0.798 $4.959 ReasoningToolsJSON Docs ↗

Capability badges appear only where the provider catalog explicitly lists support. A blank cell means the source is silent, not that the feature is absent.

Listed rates

Price across providers

Input price per million tokens as published by each provider. Bars are drawn from listed rates only — no traffic weighting, since the catalog observes no requests.

Lowest input $0.172/M

Across 18 priced providers

Median input $0.55/M

Midpoint of listed rates

Highest input $0.798/M

4.6× the lowest listed rate

Output range $1.032 – $4.959

Per million output tokens

EmpirioLabs AI $0.172/M
OrcaRouter $0.172/M
Alibaba (China) $0.172/M
Vultr $0.3/M
CrofAI $0.35/M
Kilo Gateway $0.39/M
SiliconFlow $0.39/M
TokenGo $0.4/M
Ofox $0.55/M
Ofox $0.55/M
OpenRouter $0.55/M
Jalapeno Cloud $0.6/M
LLMTR $0.6/M
Alibaba $0.6/M
DevPass (LLM Gateway) $0.6/M
Cloudflare AI Gateway $0.6/M
Cortecs $0.668/M
GreenPT $0.798/M
Cost calculator

Estimate a workload

$0.00
Excludes taxes, non-token charges, and tiered discounts.

3 providers list the identical $0.172 input rate, so price alone will not separate them — compare context limits, max output, and capabilities above.

Context limits also differ by provider, from 256K to 262.144K tokens. Compare the provider table above before choosing on price alone.

Specification

Capabilities

Recorded from the source catalog and provider listings.

Reasoning Yes
Tool calling Yes
Structured output Yes
Attachments Yes
Vision input Yes
Open weights Yes
Model family
qwen
Knowledge cutoff
Not documented
License
Not documented
Release date
2026-02-15
Model ID
alibaba/qwen3.5-397b-a17b

Weights: Hugging Face ↗

Published evaluations

Benchmarks

Every result stays attached to its source, version, metric and harness. Scores from different versions are never merged, and the profile below plots each benchmark against its own population rather than on a shared scale.

Benchmark registry
SWE-Bench Verified resolved Source ↗
SWE-Bench Verified #15 of 32

Each strip shows every published score for that benchmark, with this model marked. The lighter shape behind the ticks is the density of results, and the dashed line is the median.

Cost against capability

Price and performance

Listed input price plotted against SWE-Bench Verified, the benchmark with the widest published coverage that this model appears in.

0 51 102 $0.1 $1 $10 Qwen3.5 122B-A10B — $0.4/M, score 72.0 Qwen3.5 27B — $0.3/M, score 72.4 Qwen3.5 397B-A17B — $0.6/M, score 76.4 Qwen3.6 27B — $0.6/M, score 77.2 Qwen3.6 35B-A3B — $0.248/M, score 73.4 Qwen3.7 Max — $2.5/M, score 80.4 Claude Fable 5 — $10/M, score 95.0 Claude Opus 4.8 — $5/M, score 88.6 Claude Opus 5 — $5/M, score 96.0 Claude Sonnet 5 — $2/M, score 85.2 DeepSeek V4 Flash — $0.15/M, score 79.0 DeepSeek V4 Pro — $0.435/M, score 80.6 MAI-Code-1-Flash — $0.75/M, score 71.6 MiniMax-M2 — $0.3/M, score 69.4 MiniMax-M2.1 — $0.3/M, score 74.0 MiniMax-M2.5 — $0.3/M, score 75.8 Devstral Medium — $0.4/M, score 61.6 Devstral Small — $0.1/M, score 53.6 Mistral Medium 3.5 — $1.5/M, score 77.6 Mistral Medium (latest) — $1.5/M, score 77.6 Kimi K2 Thinking — $0.4/M, score 71.3 Kimi K2.5 — $0.3/M, score 70.8 Kimi K2.6 — $0.95/M, score 80.2 Nemotron 3 Ultra 550B A55B — $0.5/M, score 70.7 Step 3.5 Flash — $0.1/M, score 74.4 Step 3.7 Flash — $0.185/M, score 76.5 Hy3 — $0.066/M, score 78.0 Hy3 preview — $0.066/M, score 74.4 MiMo-V2.5-Pro — $0.435/M, score 78.9 GLM-4.7 — $0.6/M, score 73.8 GLM-4.7-Flash — $0.06/M, score 59.2 GLM-5 — $1/M, score 72.8 Qwen3.5 397B-A17B Input price per million tokens (log scale) Resolved

The stepped line is the efficient frontier: at each price, the best score available for that money or less. A model sitting on it is not being beaten by anything cheaper. Price is log-scaled because listed rates span four orders of magnitude. Only models with both a listed price and a score on this benchmark can appear.

Catalog activity

Change log

Field-level changes detected between successful source imports.

Full change log
Price Completion1.032 → 3.6
Price Prompt0.172 → 0.6
Provenance

Sources & verification

Every figure on this page traces back to one of these records.

Methodology
Public API

Use this record

Fetch the complete source-linked model record. No key, no account, no rate-limited tier.

API documentation
Endpoint
GET https://model.kyssta.lol/api/v1/models/alibaba/qwen3.5-397b-a17b
curl
curl "https://model.kyssta.lol/api/v1/models/alibaba/qwen3.5-397b-a17b"
Common questions

Frequently asked questions

Answered directly from the stored record — nothing here is generated beyond the catalog's own fields.

What is Qwen3.5 397B-A17B?

Large open Qwen multimodal MoE for visual agents and long technical tasks. It is published by Alibaba Qwen and catalogued here from Models.dev.

How much does Qwen3.5 397B-A17B cost?

Listed input pricing starts at $0.172 per million tokens from EmpirioLabs AI, rising to $0.798 across 18 listed providers.

What is the context length of Qwen3.5 397B-A17B?

Qwen3.5 397B-A17B accepts up to 262.144K tokens of context and returns up to 65.536K output tokens.

Does Qwen3.5 397B-A17B support tool calling and structured output?

Provider catalogs list support for tool calling, structured output, reasoning, and image input.

Which providers serve Qwen3.5 397B-A17B?

18 providers list this model: UnoRouter, EmpirioLabs AI, OrcaRouter, Alibaba (China), Vultr, CrofAI and 12 more.

Are the weights for Qwen3.5 397B-A17B open?

Yes. The weights are published and downloadable from Hugging Face.

When was Qwen3.5 397B-A17B released?

The catalog records a release date of 2026-02-15, last verified Sep 11, 2026.

More models from Alibaba Qwen

View all →