AiCostCompare

Qwen3.5 397B A17B

by Alibaba (Qwen) · qwen/qwen3.5-397b-a17b · #174 cheapest of 325 paid models

Prices updated Jul 22, 2026, 1:31 AM UTC · refreshed hourly

The Qwen3.5 series 397B-A17B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. It delivers...

Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol

Pricing

Input

$0.39

per 1M tokens

Output

$2.34

per 1M tokens

Blended (3:1)

$0.877

3 input : 1 output

Cache read

per 1M cached tokens

Cache write

per 1M tokens

Cache & batch economics

Effective prices for Qwen3.5 397B A17B (no cache-read rate published — hits billed as normal input).

Cache hit rateEffective blended $/1MExample request*vs no cache
0%$0.877$0.0421
50%$0.877$0.0421−0%
90%$0.877$0.0421−0%

*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.

What a request costs

WorkloadInput tokensOutput tokensCost / requestCost / 1K requests
Short chat message500300$0.000897$0.897
Document summary8,0001,000$0.00546$5.46
Codebase question (RAG)30,0002,000$0.0164$16.38
Long-context analysis150,0005,000$0.0702$70.20

Performance

Intelligence Index

33.7

AA composite quality

Coding Index

48.2

Math Index

Agentic Index

19.8

Output speed

61 tok/s

median

Time to first token

1.76s

median

Time to first answer

53.70s

after reasoning tokens

Artificial Analysis benchmarks

BenchmarkScore
GPQA Diamond89.3%
Humanity's Last Exam27.3%
SciCode42.0%
IFBench78.8%
AA-LCR (Long Context Reasoning)65.7%
Terminal-Bench Hard40.9%
Terminal-Bench 2.151.3%
τ²-Bench (Telecom)95.6%
τ³-Bench Banking13.4%

Design Arena Elo

CategoryElo
Website1,218
UI components1,205
Data viz1,212
SVG1,188
3D1,219
Game dev1,198
Code categories1,213

Specs

Context window

262K

tokens

Max output

66K

tokens

Input modalities

text, image, video

Tokenizer

Qwen3